Johngir & Anor v Revenue And Customs (VAT - penalty - personal liability notices) [2024] UKFTT 274 (TC) (27 March 2024)

	[Home] [Databases] [World Law] [Multidatabase Search] [Help] [Feedback]
	First-tier Tribunal (Tax)
You are here: BAILII >> Databases >> First-tier Tribunal (Tax) >> Johngir & Anor v Revenue And Customs (VAT - penalty - personal liability notices) [2024] UKFTT 274 (TC) (27 March 2024) URL: http://www.bailii.org/uk/cases/UKFTT/TC/2024/TC09120.html Cite as: [2024] UKFTT 274 (TC)

DECISION

Introduction

1. This is an appeal against HMRC's decisions to issue the appellants with personal liability notices in respect of penalties issued to Two Bros Restaurant Brands Limited (Two Bros). Both appellants were directors and shareholders of that company.

2. The company operated two restaurants (trading as Shere Khan (SK) and Oodles N'Oodles (Oodles)). The penalties arose following an HMRC VAT visit and subsequent enquiry in respect of Two Bros. HMRC contended that till data showed discrepancies and, as information and documents had not been provided, raised a VAT assessment for the periods 10/14 to 01/18 on 12 June 2018 based on the information which had been provided. The assessment was appealed to HMRC and a review was undertaken. The review reduced the assessment for one period but upheld the assessment for the other periods in full.

3. On 16 July 2018, Two Bros went into liquidation.

4. A penalty assessment was issued to Two Bros on 16 April 2019, on the basis that HMRC considered that the directors had deliberately submitted incorrect VAT returns. The penalty liability notices were issued on the same day, on the basis that the deliberate inaccuracies were attributable to the two directors equally, and because the company had become insolvent.

5. The tribunal were provided with electronic bundles totalling considerably more than 40,000 pages, as follows:

(1) main bundle (tribunal documents, correspondence, assessments, VAT returns, documents supplied by appellants) - 906 pages

(2) bundle of bank statements for May 2016 to April 2018 - 1387 pages

(3) bundle of purchase invoices - 9614 pages

(4) bundle of till reports for Oodles N'Oodles restaurant for the period 1 September 2017 to 8 November 2017 - 13380 pages

(5) bundle of till reports for Shere Khan restaurant for the period 1 September 2017 to 8 November 2017 - 15474 pages

(6) bundle of witness statements - 28 pages

Background

6. The appellants were the directors and shareholders of Two Bros, incorporated on 29 May 2014 as MW Brands Ltd. The appellants were appointed and acquired their shares on incorporation. The name was changed on 27 October 2016. The company registered for VAT with effect from 1 August 2014.

7. The company took over a restaurant business trading from two restaurants under the names 'Shere Khan' and 'Oodles N'Oodles'. These restaurants are both at the Star City leisure complex in Birmingham. The restaurants had approximately 250 covers and 60-70 employees between them. Each restaurant had two managers, two supervisors and a cashier as well as cooking and wait staff. The company records, including VAT returns, were prepared by a bookkeeper. Between 2014 and 2018 the company engaged four bookkeepers. The company entered creditors voluntary liquidation on 16 July 2018.

8. On 19 July 2017, HMRC issued a Notice of Inspection and Notice to Produce at both restaurants. The inspection was to take place between 19 and 21 July 2017. HMRC officers made an unannounced visit to both premises on 20 July 2017: the officers were told that the business objected to the unannounced visit and told HMRC to leave the premises and make an appointment to examine the business records.

9. On 21 July 2017 a further Notice of Inspection and Notice to Produce was issued, requesting a visit at Shere Khan restaurant on 25 August 2017. The Note set out the statutory records or information required to be produced at the visit. On 21 August 2017 Mr Johngir asked that the meeting be postponed as he was unable to attend and the bookkeeper was on annual leave and could not attend in his place. The meeting was postponed to 22 September 2017 and requested that the required information be provided by 21 September 2017 regardless of whether the meeting took place. On 7 September 2017 Mr Johngir asked for the meeting to be postponed again as the bookkeeper would be still away until 28 September 2017. HMRC refused to postpone the meeting further as the absence of the bookkeeper did not prevent recent records being available, as well as bank statements and annual accounts. The letter noted that the tills would also be able to be checked.

10. Following this meeting, the review continued with a meeting on 9 November 2017 at the restaurant premises and subsequently the company accountant's offices. The records and documents required were not all available. Mr Johngir confirmed that both restaurants were run by the company and were similar in size, menu, and opening hours. Each had three card machines and standalone tills. Till data was obtained by HMRC in both restaurants for the period 26 August 2017 to 8 November 2017. HMRC were advised that the tills had been changed since the initial visit on 20 July 2017. Purchase invoices, bank statements and sales information were provided at the accountant's office but no ledgers or VAT accounts were available. It was noted that sales and purchases were recorded using SAGE. On 8 December 2017 a Schedule 36 Information Notice was issued requested the sales and purchase ledgers, including SAGE reports. This information was to be provided by 8 January 2018. On 5 January 2018 the company asked for an extension of time to 8 February 2018 to provide the information. HMRC agreed that the information should be made available by 14 February 2018 and that till reviews be undertaken on 12 February 2018, with documentation reviewed at the accountant's offices from 14 February 2018. The SAGE information was noted not to have been provided.

11. Following correspondence, it was agreed that the till reviews would be undertaken on 28 February 2018. The company would provide copies of the requested documents rather than have HMRC review it at the accountant's office. It was proposed that the information be provided by the second week in April 2018. On 17 February 2018 Mr Johngir advised HMRC that the company would not be able to provide the SAGE information as they operated a manual system whereby "all input and output is calculated manually and then put in to the HMRC portal to calculate the VAT". He stated that it would be a very long task to put four years of "invoices and incomes" on to SAGE and could not agree to the costs involved.

12. On 14 March 2018 HMRC issued a further Schedule 36 Information Notice requiring information and documents and asked the company to provide a mandate to allow HMRC to approach the till provider (the tills were leased) for information to enable HMRC to check the till settings. The Information Notice required all VAT accounts, purchase invoices, expense receipts/invoices, bank statements, Z readings showing daily sales for both restaurants, statements from merchant acquirers and purchase day books for the periods 1 August 2014 to 31 January 2018. The information was to be provided by 13 April 2018.

13. On 12 April 2018 Mr Babar wrote to HMRC to advise that the company had been issued with a winding up petition by HMRC and asked that the date for compliance with the Information Notice be deferred. HMRC refused the request on the basis that the information had been requested months earlier and that, as it appeared that no attempt had been made to start to comply, no good reason for the failure had been provided.

14. On 16 May 2018 HMRC advised the company that VAT assessments would be raised, based on the information provided to date (till Z readings for most of the period 28 July 2015 to 12 September 2017; till transaction reports for the period 26 August 2017 to 13 November 2017; credit card information from merchant acquirers; annual accounts for 2015 and 2016. Bank statements and purchase invoice information had been made available for half a day at the accountant's office). A schedule of errors for each period was provided together with an explanatory letter. HMRC also advised that penalties would be raised on the basis of deliberate behaviour given the scale of disparity in cash sales between the old and new tills.

Till evidence

15. Till Z-reports were provided to HMRC for the period 28 July 2015 to 12 September 2017; transaction reports were provided for the period 26 August 2017 to 13 November 2017. HMRC compared the information in these reports to the information provided which detailed credit card sales.

16. For the periods 10/14 to 10/16, card sales amounted to £5,441,374. For the same period, gross declared sales on the VAT returns were £5,298,152. For three periods (10/14, 01/15, 10/16), card sales exceeded declared gross takings. HMRC accordingly concluded that the level of cash sales declared was not credible.

17. The Z-readings showed average daily cash sales of £1,346 across both restaurants across the period. New tills were introduced in the restaurants on 26 August 2017. For Shere Khan restaurant, the Z readings for the original till showed average daily cash takings of £824.97. The Z readings for the new till showed average daily cash sales of £2,970.78. For Oodles N'Oodles, the average daily cash takings on the original till Z readings were £645.25. On the new till, the average daily cash takings were £2,363,36.

18. The till report for Shere Khan dated 20 October 2017 showed the following information: an X reading taken at 23:41 showed 'cash in drawer' as £2,511.71. At 23:41 the till was changed to 'refund' mode and a single cash refund of £500 was entered against the 'Misc Food' category for Tab 333, with one cover. An X reading taken at 23:42 showed 'cash in drawer' as £2,011.71 and food sales similarly reduced by £500. The 'Refund/Void T1' reading showed £162.51.

19. The till report for Oodles N'Oodles dated 21 October 2017 showed similar information: an X reading had been taken at 23:48 showing the 'cash in drawer' (inter alia) as £3,401.51. At 23:49, the till was changed to 'refund' mode and a refund of £400 was recorded in a single transaction, against the 'Open Food' category, in cash. An X reading taken at 23:50 then showed the 'cash in drawer' recorded in the till report as £3,001.51. Food sales were also reduced by £400 between the two X readings. The 'Refund/Void T1' entry in the X reading showed £35.10.

20. The till report for Shere Khan dated 22 October 2017 (in the very early morning and so apparently attributable to trading on 21 October 2017) also showed similar information: an X reading taken at 00:01 showed 'cash in drawer' as £3,748.09. There was a subsequent sale recorded at 00:03, paid for by card. A nil sale was also recorded at 00:11. At 00:11 the till was changed to 'refund' mode and a single cash refund of £400 was entered for Tab 333 with one cover against the 'Misc Food' category. A Z reading taken at 00:13 showed 'cash in drawer' as £3,348.09. The 'Refund/Void T1' entry in the Z reading showed £144.80.

Validity of the assessments

21. The liquidator originally declined to appeal the assessments and penalties against the company to the Tribunal. The appellants took legal action to require the liquidator to appeal the assessment and penalties. A consent order was issued to settle that action, under which the liquidator agreed to take advice from tax counsel as to the merits of an appeal by Two Bros against the assessments, provided that the appellants paid a specific amount to cover costs.

22. The appellants did not pay that amount and so the liquidator did not pursue Two Bros' appeal against the assessments and penalties. However, HMRC accepted that the burden remains on them to demonstrate that the assessments and penalties against Two Bros were validly raised and reasonable in order to satisfy the burden of proof on them to show that that the penalty liability notices are validly raised and reasonable.

23. HMRC contended that the assessments had to be made on the basis of best judgement, as insufficient records were available to use any other method of assessment.

Output tax

24. The credit card sales information provided in the visit to the accountant's office was confirmed against the credit card information provided by merchant acquirers. For each period, the credit card sales were deducted from gross declared sales on the VAT returns to obtain the cash element of sales. This resulted in the following analysis for the periods for which credit card sales details were available:

Period Gross Declared Sales Confirmed Credit Card Sales Cash Element Calculated

10/14 £324,785.55 £557,449.75 (£232,637.42)

01/15 £403,166.50 £574,746.75 (£171,580.25)

04/15 £597,780.21 £594,934.34 £2,845.87

07/15 £579,988.06 £544,827.09 £35,160.97

10/15 £637,910.14 £628,905.35 £9,004.79

01/16 £688,140.27 £628,408.13 £59,732.14

04/16 £697,910.29 £656,923.82 £40,986.47

07/16 £751,519.92 £569,194.34 £182,325.58

10/16 £616,925.14 £685,984.75 (£69,059.61)

Totals £5,298,152.86 £5,441,374.32 (£143,221.46)

25. HMRC concluded that the sales declared did not appear to be credible and therefore reviewed the available Z readings. They had been unable to check the till settings for the original tills to ensure that all sales, voids and cancellations had been recorded correctly.

26. The transaction reports and Z readings taken from the tills on 9 November 2017 were also reviewed. For the period 26 August 2017 (when the till was installed) to 8 November 2017, the night before the visit, the till information showed gross sales across the two restaurants of £778,772.03 for the period 26 August to 31 October 2017. This was significantly more than the gross sales figure declared on the VAT return for the entire VAT period 10/17 (1 August 2017 to 31 October 2017) of £703,556.31.

27. The available information showed that for Shere Khan the average daily cash sales recorded on the original tills Z-readings were £824.97 per day. The average daily cash sales recorded on the new tills Z readings from 26 August 2017 to 8 November 2017 was £2,970.78; this was a sustained increase each day rather due to one or two anomalous readings. For Oodles N'Oodles, there was also a substantial increase in cash sales following the introduction of the new tills, as the recorded average daily cash sales increased from £645.25 to £2,363.36, also in a sustained manner. The combined average daily cash sales on the new tills amounted to £4,294.94, compared to an average of £1,346.65 on the original tills.

28. HMRC concluded that this information showed that cash sales had been suppressed.

29. An assessment to output tax was therefore made. Using the information from the new tills, it was established that card sales represented 62.38% of total sales. As the card sales information had been confirmed with the merchant acquirer data, HMRC raised the assessments on the basis that card sales for each period were 63% of gross sales.

30. For a number of periods (01/17, 04/17, 10/17 and 01/18) the credit card information was incomplete so, for these periods, the assessment was based on the suppression rate of 38.65% established for the other periods. The gross declared sales for these periods were therefore increased accordingly and the assessments calculated on that basis.

31. Mr Firth submitted that the output tax assessments were based on material which was not provided to the Tribunal: in particular, till Z reports, credit card statements detailing card sales and the annual accounts for 2016 and 2015. Further, the VAT returns alleged to contain errors for the periods 10/14, 01/15, 04/15, 07/15, 01/16 and 07/17 were also not provided to the Tribunal.

32. The Tribunal had the unchallenged evidence of an HMRC Officer as to the contents of this material, and his notes with details of figures recorded at the accountant's office were included in the bundle. It was not put to him that his evidence as to the contents of the material was incorrect. As such we consider that the information on which the assessments were based was provided to the Tribunal. Those wishing for completeness might have wanted the underlying records to be included in the bundle but, given a series of bundles which which amounted to well over 40,000 pages, the Tribunal considers that sufficient material was provided.

33. The new till was stated to have been ordered before the company was aware that HMRC were investigating their VAT position. Mr Firth contended that the change of till could not support any case of suppression; from 20 July 2017, the company and the appellants were aware of the investigation and it was submitted that it was inherently unlikely that a taxpayer would suppress turnover in circumstances where they know that HMRC are looking into the question of whether turnover has been suppressed.

34. We do not consider that this is 'inherently unlikely' as submitted; indeed, it seems probable that a taxpayer who does not intend to admit to sales suppression would seem more likely to continue the suppression, once aware of an investigation, in order to ensure that declared turnover remains consistent in the hope that the suppression remains undetected. Accordingly, we do not consider that the acquisition and use of a new till means that there cannot have been any suppression of takings.

35. It was put to Officer Beard that there were only three instances of large cash refunds shown in the till reports and these did not support a contention of suppression across the entire period, nor the amount of suppression which HMRC had contended for. Officer Beard stated that these three transactions seemed to be some sort of experiment, to see what appeared on the Z readings on the new till. He also stated that he considered that suppression was shown by the difference between the old and new till Z readings, which showed an increase of cash takings of approximately £2,000 per day after the till was replaced.

36. We find that HMRC have met the burden of proof on them to show that there were inaccuracies in the VAT returns and that they were entitled to raise assessments to output tax accordingly on the basis of best judgment.

Zero rated sales

37. The assessments were initially made on the basis that there were no zero-rated sales as no evidence had been provided to support any zero-rating, and the Z readings (on either the original or new tills) did not distinguish between standard and zero-rated sales.

38. Following discussions in ADR, HMRC reviewed the till information for three periods (1-3 September 2017; 1-2 October 2017 and 1-3 October 2017) for each restaurant to consider whether any adjustment could be made for zero-rated sales. An adjustment was calculated on the basis that relevant food in sales for those days could be zero-rated where the till transaction information showed less than 10 minutes between the meal being ordered and paid for (on the basis that these were likely to be takewaway meals). Within those sales, various items within these meals were assumed to be served cold for the purposes of calculation although they could be served hot for takeaway (such as naan bread). The most commonly used tab numbers for such meals were recorded and the rest of the till transaction information searched for those till numbers to locate takeaway meals for analysis.

39. On review, HMRC concluded that approximately 0.5% of sales were takeaway on Sunday-Thursday and that 0.7% were takeaway on Fridays and Saturdays. This was used to calculate zero-rated sales for each period and the assessments were reduced accordingly. In the hearing, the tribunal was asked to amend the personal liability notices accordingly. The appellants suggested that some of the return inaccuracies might result from zero-rated sales not being included but provided no detail in support of this suggestion.

40. The appellants contended that the assessments did not take into account zero-rated sales made by the restaurants, although the HMRC officer had accepted in his witness statement that some sales were zero-rated. Officer Beard was not cross-examined on this point, perhaps because his witness statement made it clear that, as set out above, he had revised the assessments to allow an element of zero-rated sales on the basis of information available from the till reports.

41. In their witness statements both appellants stated that "Popular take away items from Shere Khan in particular, for example items such as flavoured milkshakes, flavoured lassis, cold samosa chaats and chaats were zero rated sales" as they were for consumption off the premises. No detail was provided to show how this amounted to the zero-rated sales declared in the returns and no support for these statements was provided, although the bundle included over two months of itemised till transaction reports for Shere Khan.

42. No other evidence of zero-rated sales had been provided. The amendments were described in a letter from HMRC to the appellants of 12 March 2021. Considering the evidence put to us, we prefer Officer Beard's unchallenged evidence as to zero-rating noting that his letter, referred to in his witness statement, sets out the methodology used to analyse the transaction reports to establish a level of zero-rating in contrast to the appellants' unsupported assertions. We find therefore that the appellants' contention is not supported and that the assessments, as amended, did take into account information available as to zero-rating of supplies. We also find that the appellants have not met the burden of proof on them to displace the assessments with regard to zero-rating.

Other adjustments

43. During ADR, it appears that it was contended that the assessments should also take into account of other items:

(1) the impact of 1p sales shown in the transactions, which were stated to be recorded in order to open the till drawer rather than use the no-sale button on the till;

(2) drinks and food sales

(3) 2 for 1 offers

44. HMRC reviewed the till information and that none of these had any impact sufficient to make a further reduction in the assessments. There were no submissions in the hearing that this conclusion was incorrect.

Input tax

45. Purchase invoices were provided to HMRC at the accountant's offices, but no purchase ledgers were initially provided and the correspondence from Mr Johngir suggested that there were no such ledgers as he stated that . As such, there was no support for the amounts of input tax claimed on the VAT returns. A comparison of the VAT returns and the accounts for the business also indicated that more purchases had been claimed in the VAT returns than in the accounts.

46. As they had not been provided with support for the input tax claims, HMRC had conducted a review of purchase invoices and output tax at a restaurant operated by another company owned by the appellants (3KH) as they were advised that this operated in the same way as Two Bros. The review showed that VAT was included in approximately 20% of the total purchases. In the absence of any other information, HMRC considered that the same split would apply to the purchases of Two Bros. The information available from input tax claims and the financial accounts indicated that input tax had been claimed on 29% of purchases in 2015 and 47% in 2016. The difference between 20% and the input tax percentage claimed was disallowed in each accounting period.

47. The appellants contended that not all of the VAT returns alleged to contain errors had been provided to the Tribunal, and that HMRC had not proven that the review of purchases for 3KH had shown that 20% of invoices bore VAT. It was also contended that, even if the figure for 3KH could be proven, one would need to "go through the build up of the respective companies' VAT to establish what, if any, comparisons could properly be drawn". HMRC's evidence was that they had been advised that the 3KH restaurant and company operated in a very similar way to Two Bros and these two restaurants. This was not challenged, and neither of the appellants stated that this was incorrect.

48. The Tribunal bundle contained copy VAT returns each followed by a list of purchases. The appellants submitted that ledgers had therefore been provided showing the build up of input VAT, and that these ledgers added up to the amounts in the VAT return. This assertion was not supported by any detailed submissions or evidence showing how the purchase invoices provided to the Tribunal supported the lists of purchases. For example, in the lists of purchases there are significant amounts of input tax stated to have been paid in respect of supplies of staff and rent. We were not taken to any invoices in respect of these.

49. From the correspondence, it appears that these lists may have been provided as part of the ADR process and a letter from HMRC dated 12 March 2021, written following the ADR meetings, refers to these lists and amends the 01/16 and 10/17 periods to reduce the assessments. The analysis for the 01/17 period indicated that the amount disallowed following this review should be higher than that originally assessed, but HMRC decided not to increase the assessment.

50. That letter also notes that ledgers had not been provided for the 10/14, 01/15, 04/15 and 07/15 periods. HMRC had reviewed the amounts for these periods on the basis of the allowed input tax for periods claimed; the review disallowed amounts were higher than the amounts disallowed in the original assessments and HMRC decided not to amend those assessments. Officer Beard's witness statement noted that the input tax assessments had been amended, based on additional purchase invoices provided during ADR.

51. Following ADR, HMRC were provided with additional material in respect of input VAT and Officer Beard's unchallenged evidence was that this was used to amend the input tax assessments. As set out in HMRC's letter of 12 March 2021, the evidence was reviewed and input tax allowed where the evidence supported the claim. Where the amendments resulted in higher disallowances, no amendment was made. For periods where ledgers were missing, the rate of input tax allowed for the other periods reviewed (37.24%) was used to recalculate the assessments. Where, again, this resulted in a higher amount disallowed, the assessment was not varied. This evidence was not challenged.

52. To the extent that the appellants' submissions and evidence were intended to assert that the amounts declared as input tax on the return were reliable or that HMRC's assessments had failed to take evidence into account, we cannot agree. The information in the purchases lists has not been shown to correlate closely to invoices provided and the HMRC letter of 12 March 2021 confirms Officer Beard's evidence that the assessments were reviewed to take into account the information provided.

Conclusion as to assessments

53. s73(1) of the Value Added Tax Act provides that

"Where a person has failed to make any returns required under this Act (or under any provision repealed by this Act) or to keep any documents and afford the facilities necessary to verify such returns or where it appears to the Commissioners that such returns are incomplete or incorrect, they may assess the amount of VAT due from him to the best of their judgment and notify it to him."

54. The meaning of 'best judgment' is set out in Van Boeckel v Customs and Excise Commissioners [1981] STC 290:

"What the words 'best of their judgment' envisage, in my view, is that the Commissioners will fairly consider all material placed before them and, on that material, come to a decision which is one which is reasonable and not arbitrary as to the amount of tax which is due. As long as there is some material on which the Commissioners can reasonably act then they are not required to carry out investigations which may or may not result in further material being placed before them" (at p 292)

55. Further guidance was given in Carnwath J in Rahman (t/a Khayam Restaurant) v CEC [1998] STC 826 (at 835):

"... there are dangers in taking Woolf J's analysis of the concept of "best judgment" out of context ... the tribunal should not treat an assessment as invalid merely because it disagrees as to how the judgment should have been exercised. A much stronger finding is required; for example, that the assessment had been reached "dishonestly or vindictively or capriciously"; or is "spurious estimate or guess in which all elements of judgment are missing"; or is "wholly unreasonable". In substance those tests are indistinguishable from the familiar Wednesbury principles (see Associated Provincial Picture Houses Ltd v Wednesbury Corp [1948] 1 KB 223). Short of such a finding, there is no justification for setting aside the assessment."

56. In effect, in exercising best judgment an HMRC officer is simply required not to be arbitrary or to guess, he must not act from wrong motives, and he is required not to act wholly unreasonably. But he is not required to be as right as it is possible to be.

57. The position was also confirmed in in Customs and Excise Commissioners v Pegasus Birds Ltd [2004] EWCA Civ 1015 which then cautioned against allowing an appeal routinely to become an investigation of the bona fides or rationality of the "best of judgment" assessment made by Customs:

"The tribunal should remember that its primary task is to find the correct amount of tax, so far as possible on the material properly available to it, the burden resting on the taxpayer. In all but very exceptional cases, that should be the focus of the hearing, and the Tribunal should not allow it to be diverted into an attack on the Commissioners' exercise of judgment at the time of the assessment." (para 38(i))

58. We find that the appellants failed to keep documents necessary to verify returns; for example, there are clearly purchase records missing in respect of substantial amounts of input tax.

59. We also consider that it was reasonable for HMRC to consider that the returns were incorrect in respect of output tax, given that recorded card sales exceeding declared turnover for certain periods and the difference in cash sales recorded on the introduction of the new tills. In respect of the periods in which the new tills were used and for which credit card information was not confirmed, we note that the declared gross sales remained significantly below the information recorded by the tills in the period for which the new till information was available such that it was reasonable to conclude that sales were still being suppressed.

60. On the evidence before us and bearing in mind the case law, we find that the assessments were made to best judgement. There was no submission that the assessments were made arbitrarily or otherwise incorrectly motivated, and we do not consider that there are any grounds to indicate that HMRC acted wholly unreasonably in making the assessment. The assessments for these periods were based on the material available to HMRC and have been amended to take into account further material later provided, and we consider that they amount to a reasonable assessment of the tax due in the circumstances.

61. In the hearing there appeared to be a suggestion that HMRC needed to prove that the amounts assessed as inaccurate were correct rather than to best judgement. No support was provided for this but, to the extent that this was intended to indicate that something other than 'best judgement' was required to be shown, we consider that it is well-established law that an assessment to VAT stands good unless the appellant is able to produce evidence to show that it is wrong; the fact that this is an appeal against personal liability notices which derive from the assessment cannot alter that.

62. As we have concluded that the assessments were made to best judgement, the burden of proof moves to the appellants to show what the correct liability should be. For the reasons set out above, we do not consider that the appellants have shown on the balance of probabilities that the assessments, as amended, are incorrect.

Penalties

63. For the appellants, Mr Firth contended that there had been no proper pleading of the allegations of deliberate behaviour in HMRC's Statement of Case. On the basis that this was a serious allegation, tantamount to dishonesty, it was contended that the rules and principles relating to such allegations applied and had not been met in this case.

64. Such allegations needed to be properly pleaded and particularised in advance, as set out in Three Rivers District Council v Governor and Company of the Bank of England [2001] UKHL 16 ('Three Rivers'). This concluded that a defendant is entitled to know the case that they must meet and requires details of the primary facts relied on to justify the inference. Inferences could not be made from unpleaded facts.

65. The Statement of Case states the following (sic):

"HMRC submit that ... the inaccuracies in Two Bros' VAT returns were attributable to deliberate but not concealed behaviour ... HMRC submit that, as the directors who ran the business, Mr Babar and Mr Johngir must have known that they were not recording all cash sales, they had no evidence to substantiate zero rated sales [the Statement of Case subsequently sets out the revised position after material was provided following ADR, modifying this statement] and, that they had claimed input tax in excess of what was evidenced. HMRC submit that it is clear that inaccuracies on such a scale could not have arisen from mere oversight."

66. The Statement of Case also sets out why HMRC considered that cash sales were not all recorded, the lack of evidence as to zero rating, and the lack of evidence in respect of input tax.

67. Mr Firth contended that this was inadequate, as there was no particularisation of the primary facts. Mr Firth cited the commercial court decision in Jinxin Inc v Aser Media Pte Ltd and others [2022] EWHC 2988 (Comm) at [41], that all primary facts on which an inference of fraud is drawn must be pleaded. For example, he contended that HMRC had to plead what the appellants were alleged to have known, why and the basis on which the inference of deliberate behaviour was made. He considered that they had not, for example, stated in respect of the cash sales "recorded where, in what periods". This appears to overlook the sections of HMRC's Statement of Case which set out why HMRC considered that there had been suppression of cash sales and for which periods.

68. To the extent that this was intended to mean that HMRC should have been more specific, such that they should have said (for example) in their submission "that they were not recording all cash sales in the turnover declared on VAT returns for periods X-Y", we note that Gerko v Seal and others [2023] EWHC 63 (KB) states that a "concise summary" is required, and notes that the Queen's Bench Guide provides guidelines which should be followed as good and proper practice [and] that "a statement of case must be as brief and concise as possible and confined to setting out the bald facts and not the evidence of them" (at [25], quoting with approval from Portland Stone v Barclays [2018] EWHC 2341).

69. We therefore do not agree with Mr Firth that HMRC is required to set out such detail supporting the alleged facts nor do we consider that it is necessary for HMRC to repeat in such a summary the information that is set out elsewhere in the Statement of Case (such as, for example, the VAT returns involved).

70. We considered the decision in Danapal [2023] UKUT 86 (TCC), also cited by Mr Firth, and note that the Upper Tribunal in that case concluded that the First tier Tribunal (FTT) should not have made findings of dishonesty against a firm which had not been included in pleadings nor put to the relevant witness in cross-examination. However, in that case, it appears that HMRC did not in fact make any allegations of dishonesty against that firm (see [48] and [49]); this was not an instance of HMRC failing to adequately plead dishonesty in their Statement of Case but appears rather to have been a conclusion reached by the FTT on its own account. We consider that it is therefore of little assistance in this case.

71. As summarised in Gerko at [36] "The court and the defendants to a pleading of dishonest conspiracy are entitled to expect a clear concise statement of relevant facts, where the key allegations of dishonesty and the acts done in furtherance of the conspiracy are clearly set out with sufficient relevant particulars of primary fact".

72. Bearing this in mind, in our view, the Statement of Case is sufficiently particularised. It makes it clear that deliberate behaviour is alleged against the appellants and sets out the key facts on which HMRC relied: that the appellants were the directors running the the business, that cash sales were not all included in the turnover in VAT returns, and there was insufficient evidence regarding zero rated sales to support the returned information, and that they did not have the evidence required to support the input tax claimed in the returns.

73. Mr Firth contended that it was unclear what "ran the business" should be taken to mean, given that the restaurants had managers, cashiers and bookkeeper. This is the sort of detail which Gerko makes clear does not need to be included in allegations of dishonesty: it is a straightforward English phrase and we consider that it does not need detailed exposition to be understood in context.

74. It was submitted that HMRC had introduced an unpleaded allegation that suppression was carried out by restaurant managers acting on the instruction of the appellants, and that HMRC were therefore not even advancing the allegation that it was the appellants actually carrying out the suppression.

75. This latter submission seems somewhat misguided; if the managers were acting on the instructions of the appellants then the appellants would, in our view, be carrying out the suppression. The use of a hammer to hit a nail does not mean that the person wielding the hammer has not driven the nail into a wall. We do not consider that HMRC were making allegations of dishonesty against the managers in this submission, nor do we make any findings in respect of the quality of the behaviour of unidentified managers of the restaurants.

76. With regard to the submission as to the unpleaded allegation, we consider that this is not made out. The reference in the skeleton argument to the managers was based on information in both the appellants' witness statements and Officer Beard's witness statement, all of which were produced after the Statement of Case. It is an inference from evidence and, even if all the evidence had been available prior to the Statement of Case, it is again, in our view, the sort of detail which Gerko concluded did not need to be included in the summary, rather than a separate allegation.

77. We note also that the Statement of Case was produced before ADR. It also seems to us inevitable that, where appellants do not provide detailed explanations to HMRC and do not provide detailed grounds of appeal, the level of detail which Mr Firth appears to consider required is unlikely to be available at the point at which HMRC has to produce a Statement of Case (noting in particular that this is before witness statements are exchanged). It cannot be correct that allegations of deliberate behaviour in respect of such appellants would have to fail due to lack of proper pleading and particularisation.

78. Finally, and in support of this, we note also that "in view of the common feature of fraud claims that the Defendant will, if the underlying allegation is true, have tried to shroud his conduct in secrecy, the Court should adopt a "generous" approach to pleadings" (Gerko at [35]).

79. For these reasons, we find that the allegations of deliberate behaviour were properly pleaded and/or particularised.

Validity of the penalties - whether inaccuracies established

80. For the appellants, Mr Firth contended that HMRC had not met the burden of proof upon them with regard to the alleged inaccuracies.

81. We do not agreed, as noted above. It is clear from the evidence of Officer Beard that, for example, card sales exceeded declared turnover in various periods, that there was insufficient evidence to support input tax claims and that there were other inconsistencies which indicated that the VAT returns were inaccurate. We find therefore that there were inaccuracies in the relevant VAT returns. The question is then whether these were deliberate inaccuracies and whether the relevant behaviour was attributable to the appellants.

Whether inaccuracies were deliberate

82. The test of deliberate inaccuracy was set out by the Supreme Court in Tooth [2021] UKSC 17: the taxpayer must know that the return contains an inaccuracy and intend HMRC to be misled by the inaccuracy. HMRC agreed that deliberate behaviour required an intention to mislead HMRC.

83. HMRC submitted that the evidence of the directors was that they had overall control of the business and were responsible for the day to day running of the restaurants, visiting frequently, and working from the back office on site for periods. They reviewed weekly sales information and had oversight of purchase invoices and checked these against delivery notes. They knew the cashing up procedure and confirmed that cash was not always banked, and had described their behaviour as hands-on rather than strategic.

84. The information obtained from the new tills indicated that there had been similar suppression at separate restaurants at approximately the same time and date. It was contended that this showed that there was a controlling mind instructing persons at the restaurant to undertake the actions; HMRC contended that this was, on the balance of probabilities, deliberate behaviour by the directors, and suggested that the suppression was undertaken by managers acting on the appellants' instructions at the restaurants.

85. HMRC further contended that the appellants must have known that not all sales were recorded in the VAT returns and that the input tax claimed was in excess of the amounts for which evidence was available. The scale of the inaccuracies was such that this could not be mere oversight.

86. The appellants submitted that HMRC had not set out an adequate case to show that each of the appellants knew what was in the VAT return and knew that it was incorrect but still proceeded to submit the return to HMRC.

Discussion

Cash sales inaccuracies

87. Neither appellant accepted that cash sales were suppressed. In cross-examination, however, Mr Babar stated that he thought that £2,845 declared in one quarter, which amounted to £31 per day cash takings in that period, seemed "a bit low". Mr Johngir, asked whether he thought this amount was accurate for the period, said that he didn't know and this was the bookkeeper's work, although he thought there was very little cash taken in Oodles N'Oodles.

88. The appellants evidence was that they had no knowledge of the three large refunds shown in the till transactions on 20-22 October 2017. Mr Johngir stated that it appeared that a cashier was logged in. He denied that anyone would have taken cash out of the till on his instructions. Mr Babar said that these looks like a refund of some sort and that it was the first time he had seen it. He did not know what it could be, although he thought something was clearly wrong. He stated that neither he nor Mr Johngir ever handled the till.

89. The change in cash sales shown in the till information on the change of tills is substantial, with the Z-readings showing that cash takings increased by approximately £2,000 per day for Shere Khan and £1,700 per day for Oodles N'Oodles. That is an overnight increase of approximately 360% in recorded daily cash takings for each restaurant on the change of till.

90. Mr Johngir stated in his witness statement that "In my role as a director, I would describe myself as hands on rather then (sic) strategic." Mr Babar similarly stated that "In my role as director, I would describe myself as hands on rather then (sic) strategic." (Both at [14] of the respective witness statements).

91. Both appellants stated in their witness statements that they took the Z-readings each week to the bookkeeper. They described the process of recording cash purchases by the restaurants on the Z-readings and stated that the Z-readings would usually match the handwritten figures. They stated that Mr Johngir had "weekly reviews with the managers when the sales recorded on the Z reports were looked at". Mr Johngir stated that he "would carry out a task in performance reviews by calculating sales less purchases to generate a margin. Then [he] would deduct wages and fixed overheads such as utilities, rent and rates leaving a net figure that would indicate if the business were in positive".

92. In cross-examination Mr Johngir said that they did not review the Z-readings and instead relied on the bookkeeper to give them an overview of sales and the performance of the restaurants. Mr Johngir stated that the bookkeeper would tell them what information was needed and the managers would provide what he asked them for, including the Z-readings. Mr Babar stated that the bookkeeper would give a position of how the business was performing, with day to day sales and purchases.

93. We do not consider that the appellants' evidence in the hearing was credible, considering the inconsistencies with their clear witness statements. We find that the appellants were familiar with the contents of the Z-readings around the time that they were taken. We conclude that they would have been aware of the substantial change in recorded cash takings when the new tills were installed. There was no evidence that appellants took any action in respect of this substantial change and as such we consider that they knew and intended that cash takings were not being properly recorded in the tills.

94. Given that, we find that the appellants deliberately suppressed cash sales in the business and deliberately provided incorrect information to the bookkeeper, intending that the inaccurate information would be used in VAT returns submitted to HMRC.

Zero-rating

95. The appellants submitted that HMRC's case with regard to zero-rated sales failed as HMRC had subsequently accepted that there were zero-rated sales.

96. With regard to the zero-rating inaccuracies, we do not agree with the appellants' submission that as HMRC now accept that there were some zero-rated sales that their case regarding deliberate behaviour with regard to the zero-rated sales inaccuracies must fail. It is not a binary matter; the fact that there are some zero-rated sales does not mean that the amount of zero-rated sales declared in the returns must be treated as accurate or not deliberately mis-declared. As noted above, we find that the VAT returns did contain inaccuracies with regard to zero-rated sales.

97. The VAT returns show that the company claimed that 14% of total sales were zero-rated. An analysis by HMRC once documents were made available estimated that approximate 0.5% to 0.7% of sales were zero-rated. Officer Beard was not challenged as to this. The appellants' assertions that there were substantial amounts of zero-rated sales was unsupported; we were not taken to any supporting evidence. The appellants' witness statements show that the appellants understood the type of sales that could be zero-rated, as noted above. As the appellants' witness statements state that they provided details of takings to the bookkeeper so that the bookkeeper could complete the VAT returns, we find that the appellants deliberately provided inaccurate information as to zero rated sales knowing that it would be used in the VAT returns submitted to HMRC.

Input tax inaccuracies

98. The appellants' witness statements confirmed that they both oversaw purchase invoices and ensured that they matched delivery notes. In the hearing Mr Babar stated instead that this was the managers' responsibility and he simply overlooked their work, and he took their work on trust rather than checking the figures. Mr Johngir stated in the hearing that he did not pass any invoices to the bookkeeper that were for any purchaser other than the company, Two Bros. Officer Beard's revised assessment of input tax following ADR, referred to in his witness statement, was not challenged.

99. As above, considering these inconsistencies, we do not consider that the evidence in the hearing was credible and we consider that the appellants knew that they were providing the bookkeeper with invoices which were not addressed to the company with the intention that the VAT on these invoices would be reclaimed such that the VAT returns would be inaccurate.

Completion of the returns

100. The appellants also submitted that HMRC had failed to show that they were aware that there were inaccuracies in the VAT returns, and stated that the VAT returns had been prepared and submitted by the bookkeeper.

101. As noted above, we find that the appellants were each aware of the information that was passed to the bookkeeper and from which the VAT returns were prepared and we find that they knew that this information did not correctly reflect the sales made by the business. We find that they knew that this information would be used in completing VAT returns to HMRC which would be incorrect. We therefore find that the behaviour which led to the inaccuracies in the VAT return was deliberate behaviour by the appellants.

Conclusion

102. As we have found that the appellants' behaviour led to the inaccuracies, and that this behaviour was deliberate, we find that the penalties and personal liability notices were validly raised. We consider that the decision to transfer the penalty to the appellants on a 50/50 basis was appropriate and reasonable.

103. With regard to quantum, HMRC submitted that the penalty and so the personal liability notices should be reduced to reflect the amendments made by Officer Beard following ADR. The VAT assessments were reduced from a total of £1,321,268 to £1,187,551 and we find that the penalty and personal liability notices should be similarly reduced.

104. HMRC assessed the penalty on the basis that disclosure was prompted and gave reduction of 5% for telling, 10% for helping and 10% for giving information. The penalty was therefore 61.25%. We do not consider that there are any grounds for amending the mitigation of the penalty. HMRC also concluded that there were no special circumstances which would merit any further reduction in the penalty. We consider that this was a reasonable conclusion.

105. The penalty is therefore 61.25% of £1,187,551, being £727,374.98. This is to be transferred equally to the appellants.

106. We therefore uphold the personal liability notices in respect of both of the appellants in the amended amount of £363,687.49 each. The appeal is dismissed.

Right to apply for permission to appeal

107. This document contains full findings of fact and reasons for the decision. Any party dissatisfied with this decision has a right to apply for permission to appeal against it pursuant to Rule 39 of the Tribunal Procedure (First-tier Tribunal) (Tax Chamber) Rules 2009. The application must be received by this Tribunal not later than 56 days after this decision is sent to that party. The parties are referred to "Guidance to accompany a Decision from the First-tier Tribunal (Tax Chamber)" which accompanies and forms part of this decision notice.

ANNE FAIRPO

TRIBUNAL JUDGE

Release date: 27^th MARCH 2024

First-tier Tribunal (Tax)