[2014] UKFTT 632 (TC)

[image removed]

TC03756

 

 

 

Appeal number: TC/2012/06330

 

VALUE ADDED TAX – assessments – best judgment – examination of till records – programming of till – whether transactions fully reflected in VAT returns – evidence showing omission of transactions from main records – assessments confirmed – penalties under s 60 VATA 1994 and Sch 24 FA 2007 also confirmed – appeals dismissed

 

 

FIRST-TIER TRIBUNAL

TAX CHAMBER

 

	SERLA LTD	Appellant
	- and -
	THE COMMISSIONERS FOR HER MAJESTY’S	Respondents
	REVENUE & CUSTOMS

 

 

TRIBUNAL:	JUDGE  JOHN CLARK
	MR PHILIP GILLETT FCA

 

 

 

Sitting in public at 45 Bedford Square, London WC1B 3DN on 1 and 2 April 2014

 

 

Ahmet Mustafa Osam, Legal Department, Zek & Co, Accountancy Services, for the Appellant

 

Jake Hillier, HM Revenue and Customs Local Compliance Appeals & Reviews, for the Respondents

 

 

 

 

© CROWN COPYRIGHT 2014

DECISION

 

 

1.             The Appellant (“Serla Ltd”) appeals against assessments to VAT in respect of the VAT periods ending from 31 August 2008 to 30 November 2011, against a civil evasion penalty assessment in respect of the VAT periods ending between 31 August 2008 and 28 February 2009, and penalty assessments in respect of the VAT periods ending between 31 May 2009 and 30 November 2011.

The background facts

2.             The evidence consisted of a bundle of documents, which included a series of witness statements. On behalf of Serla Ltd, statements were given by Mr Ali Duzgun, the director of Serla Ltd, and by Mr Zekai of Zek & Co. Although a statement had been given by Mr Cem Duzgun, Serla Ltd did not rely on it as Mr Cem Duzgun was not available for cross-examination because he was away in Turkey. On behalf of the Respondents (“HMRC”), evidence was given by Mr Kalpesh Patel, Mr Christopher Nowak, Ms Janet Harter, officers of HMRC, and Ms Georgina Arridge, a Higher Officer of HMRC. All the witnesses gave oral evidence. From the evidence we find the following background facts; where appropriate, we consider other issues of fact later in this decision.

3.             Although there was no formally agreed statement of facts, the following facts set out in HMRC’s Appendix 1 to their Amended Statement of Case were apparent from information held by HMRC and from documents held by Serla Ltd. We are satisfied that this account of the facts is correct and consistent with the remainder of the evidence. Where appropriate, we have adapted HMRC’s informal statement of facts to take account of other matters contained in the evidence. From the evidence we find the following background facts.

4.             Serla Ltd registered for VAT with effect from 5 May 2008, after taking over the business as a “transfer of going concern”. The activities of the business are those of a supermarket/general store which sells groceries, alcohol, tobacco, fruit and vegetables and general goods. The premises are open seven days a week.

5.             One electronic till is used in the shop. Till readings are obtained on a daily basis, usually by Mr Ali Duzgun, the director, who records the figures from the till reading on a daily takings sheet provided by the accountant, Zek & Co. The completed daily takings sheets and supporting till readings are provided to Zek & Co every quarter, along with the purchases and wages records, for preparation of the VAT return.

6.             The business operates HMRC’s retail scheme ‘Apportionment Scheme 1’ for the calculation of its VAT. This Scheme is based on the value of purchases for retail sale at different rates of VAT and then applying the proportion of those purchases to the sales and calculating the output tax due accordingly.

7.             On 7 October 2010, HMRC Officers Georgina Arridge and Kalpesh Patel carried out an unannounced VAT visit to the business premises of Serla Ltd. Various reports were obtained from the electronic till, as well as used journal till rolls. A questionnaire was completed with Mr Candemir, the Manager of the business, as well as Mr Ali Duzgun by telephone.

8.             On 21 October 2010, a second unannounced visit to the business premises took place. Again, the HMRC Officers involved were Ms Arridge and Mr Patel. Mr Ali Duzgun was present. Available till reports were collected and additional till interrogations were performed.

9.             Following a telephone conversation, Ms Harter wrote to Serla Ltd on 17 January 2011 to confirm that a meeting between HMRC and Serla Ltd had been arranged at Zek & Co for 15 February 2011. A schedule of the books and records Ms Harter wished to see was enclosed with that letter.

10.         On 15 February 2011 the arranged meeting took place between Mr Ali Duzgun, Zek & Co, Ms Harter and Ms Liz Sellick, another HMRC Officer.

11.         On 1 June 2011, Ms Harter and HMRC Officer Ms Mandy Sleight carried out a third unannounced visit to the business premises. Mr Ali Duzgun was present. A till interrogation was performed and till reports were printed.

12.         On 22 June 2011 Ms Harter sent Zek & Co a letter enclosing copies of the till reports for 30 and 31 May 2011.

13.         On 18 July 2011 Ms Harter sent Serla Ltd a letter requesting a meeting on 23 August 2011 and enclosing a copy of HMRC’s Public Notice 160 (Compliance checks into indirect tax matters).

14.         On 23 August 2011 the proposed meeting took place between Mr Ali Duzgun, Zek & Co, Ms Harter and Mr Nowak. The meeting was adjourned as Mr Duzgun indicated that he did not understand Public Notice 160 and needed further time for consideration. Mr Duzgun provided a signed witness statement to HMRC.

15.         On 5 September 2011, Ms Harter sent Serla Ltd a letter requesting a further meeting on 29 September 2011. That meeting took place on the latter date, the persons present being Mr Ali Duzgun, Zek & Co, Mr Osam as Mr Duzgun’s legal representative, Ms Harter, Mr Nowak and Ms Sellick.

16.         On 28 October 2011 Zek & Co wrote to Ms Harter. This letter explained that, while Mr Duzgun refuted any accusation of dishonesty, he accepted that human error may have contributed to an incorrect till reading on 2 June 2011. The letter also provided an explanation for any inconsistency in the audit report for test purchases.

17.         Copies of the notes of the meeting on 29 September 2011 were posted to Serla Ltd on 7 November 2011; HMRC invited Serla Ltd to sign and return the additional copy, noting any comments or suggested amendments as it so wished. No signed copy of the notes was received by HMRC. Also on 7 November 2011, Ms Harter wrote to Zek & Co regarding the points raised in their letter of 28 October 2011.

18.         Zek & Co replied to Ms Harter on 27 November 2011 noting her comments and relaying Mr Duzgun’s suggestion ‘that an invigilation may offer a satisfactory way forward’. As Zek & Co had not received a reply to that letter, they sent Ms Harter a reminder on 2 February 2012. Ms Harter replied on 15 February 2012, and explained that she had not received the earlier letter from Zek & Co. She requested a copy of the Z reading taken subsequently to the Z reading which she had taken on 1 June 2011. She explained that she did not consider invigilation to be appropriate, as she had obtained sufficient information from the prime records of Serla Ltd; she proposed to raise an assessment for the under-declared sales, the total amount of the assessment being £14,617.00.

19.         On 12 March 2012 Zek & Co sent Ms Harter a copy of the till reading requested by her in relation to 1 June 2011.

20.         Ms Harter wrote to Serla Ltd, for Mr Duzgun’s attention, on 22 March 2012. She acknowledged receipt of the till reading, and enclosed copies of Notice FS9 Human Rights, and Notice FS12 Compliance checks – Penalties for VAT and Excise wrongdoing. She referred to her letter dated 18 July 2011 (giving the incorrect year 2012) with which she had enclosed Notice 160, and stated that Mr Duzgun had confirmed during the meeting on 29 September 2011 that he had read and understood this Notice.

21.         On 3 April 2012 an assessment under s 73(1) of the Value Added Tax Act 1994 (“VATA 1994”) was raised in the sum of £14,617 VAT and £681.11 interest, which covered the period from 1 June 2008 to 30 November 2011.

22.         Zek & Co wrote to Ms Harter on 14 April 2012 stating that they were unable to agree her calculations in relation to the assessment raised and suggesting a further meeting in May 2012, the purpose being ‘to find and reach an agreed figure based on reasoned and explicable grounds’.

23.         On 18 April 2012 Ms Harter replied to Zek & Co, stating that HMRC ‘do not negotiate VAT settlements’. She gave further details of discrepancies. She reiterated the information concerning the right of appeal, including the offer of a review of the decision to raise the assessment.

24.         On 8 May 2012, HMRC issued to Serla Ltd a Notice of Penalty Assessment under Schedule 24 to the Finance Act 2007 (“Sch 24 FA 2007”) and Penalty Explanation schedules for the VAT periods from 1 March 2009 to 30 November 2011, for the total sum of £10,686.20 (“the Sch 24 Penalty”).

25.         On 20 June 2012, Ms Harter notified Serla Ltd of a penalty under s 60 VATA 1994 in the sum of £1,635 for tax periods 08/08, 11/08 and 02/09, ie the total period from 1 June 2008 to 28 February 2009 (“the s 60 Penalty”). In this letter she confirmed that a reduction of 20 per cent had been given for the co-operation received. Her letter also contained an offer of a review and gave details of how an appeal to the Tribunal could be made.

26.         On 12 June 2012 Zek & Co gave Notice of Appeal to HM Courts & Tribunals Service (HMC&TS”) against the assessment dated 3 April 2012, the Sch 24 Penalty and the s 60 Penalty.

27.         On 15 November 2012 Ms Harter wrote to Serla Ltd. She explained that the assessment had been miscalculated and understated the amount of additional tax due. As a result, she intended to raise a further assessment which was two years after the end of the prescribed accounting period. She referred to s 73(6)(b) VATA 1994. She enclosed a schedule showing details of the calculation of the additional tax due, the total being £4,101.00.

28.         This further assessment (“the Further Assessment”) was raised by HMRC on 22 November 2012 under s 73(1) VATA 1994 in the sum of £4,101.00 plus interest. This increased the assessed amount disputed between the parties from £14,617 to £18,718, plus interest.

29.         On 19 December 2012 HMRC issued a penalty explanation with a penalty schedule detailing penalties totalling £3,484.85 (referred to together as “the Additional Penalty”) which HMRC intended to charge Serla Ltd. These penalties were to be raised under Sch 24 FA 2007.

30.         On 31 January 2013 HMRC issued a penalty assessment notice in respect of the Additional Penalty.

31.         As the latter penalty assessment gave incorrect details of tax periods, a revised notice of penalty assessment in respect of the Additional Penalty was issued on 13 February 2013 to replace the incorrect penalty assessment notice.

32.         As at the date of the hearing, no appeals had been made against the Further Assessment or against the assessment in respect of the Additional Penalty. We refer to this below.

Preliminary issues

33.         At an earlier stage in the appeal process, Mr Hillier replied to HMC&TS on 24 May 2013 setting out the details of the matters under appeal. He indicated that HMRC had no objection to the appeal against the VAT assessment and the Sch 24 Penalty proceeding out of time. He explained that the Further Assessment and the Additional Penalty had been issued after Serla Ltd had lodged with HMC&TS its Notice of Appeal against the assessment dated 3 April 2012, the Sch 24 Penalty and the s 60 Penalty.

34.         He continued:

“To the Respondents’ knowledge, no formal appeal has been lodged either with HMRC or the Tribunal in respect of the assessment at 2a or the penalty notice at 2b. It is clear however that both of these are in dispute.

The parties wish the Tribunal to treat the assessment at 2a and the penalty notice at 2b as having been notified and dealt with as part of the existing appeal. Absent of any formal notice of appeal being given against these decisions, should the Appellant:

• now make an appeal against those decisions and

• apply for permission to appeal out of time and

• ask that the appeal against those decision be joined and dealt with as part of the existing appeal

then the Respondents would not object to any of these applications.”

35.         No action was taken by Serla Ltd or its advisers in response to that letter either in 2013 or at any later point up to the hearing of the appeal.

36.         At the beginning of the hearing, we explained that in the absence of a properly notified appeal against the Further Assessment and the Additional Penalty, we would have no jurisdiction to consider them, and that whatever the outcome of the appeal against the other matters, the Further Assessment would remain finally determined in its original amount and the Additional Penalty would remain due as determined.

37.         Mr Hillier referred to his letter of the previous 24 May and reiterated that HMRC would not object to an appeal being lodged to enable these matters to be considered by the Tribunal.

38.         We decided that it would be inappropriate to adjourn the existing appeal in order to enable the further appeal to be lodged, and that it would be in the interests of justice for Serla Ltd to be permitted to notify such an appeal as soon as possible after the end of the hearing. In the circumstances, we agreed that such an appeal could be made out of time. We also decided that the further appeal should be consolidated with the existing appeal. Mr Hillier confirmed that HMRC did not object to these arrangements. On that basis, we agreed to proceed to hear the matters covered by the existing appeal and also those to be included in the further appeal.

39.         After the hearing, an application dated 12 April 2014 was sent to the Tribunal on behalf of Serla Ltd. This application sought to amend the Notice of Appeal filed on 12 June 2012 by adding details relating to the Further Assessment and the Additional Penalty and by adding further matters to the grounds of appeal.

40.         On 22 April 2014 Mr Hillier wrote to the London office of the First-tier Tribunal (Tax) with HMRC’s comments on the application. HMRC did not object to the first part of the application, but objected to the second, as the additional grounds of appeal sought to introduce further argument after a hearing had been held on the matter.

41.         On 24 April 2014 the Judge gave Directions that Serla Ltd should within seven days give Notice of Appeal in respect of the Further Assessment and the Additional Penalty; the grounds of appeal should be the same as those for the original appeal. On condition of compliance with the latter direction, the further appeal was to be consolidated with the original appeal. The application to amend the grounds of appeal was refused.

42.         On 30 April 2014, Zek & Co responded.by sending to the First-tier Tribunal (Tax) London office a “Notice of Consolidation Amendment”. This again sought to amend the original Notice of Appeal.

43.         On 13 May 2014 the London office wrote to Zek & Co to explain that in order to comply with the Direction, it was necessary for Serla Ltd to complete and submit a new Notice of Appeal form to the HMC&TS Birmingham office, showing the same grounds of appeal as set out in the original Notice of Appeal. The document which had been sent on 30 April 2014 did not comply with the Direction.

44.         As the decision was still in the course of preparation and as HMRC had previously indicated that they would not object to the late lodging of a Notice of Appeal in respect of the Further Assessment and the Additional Penalty, the Judge had decided to grant Serla Ltd a further seven days to comply properly with the Direction by lodging a Notice of Appeal with the HMC&TS Birmingham office setting out the identical grounds of appeal. The Judge had also stated that failure to comply with the Direction within the extended period now being allowed would mean that the Tribunal would be compelled to deal with its decision on the basis that the Further Assessment and the Additional Penalty were not under appeal and therefore stood as made.

45.         On 19 May 2014 Zek & Co sent the further Notice of Appeal to HMC&TS Birmingham; this was acknowledged by that office on 22 May 2014.

46.         We have been notified that the further appeal has been lodged under the reference TC/2014/02776; as the Directions have now been complied with, that appeal is consolidated with the original appeal under reference TC/2010/06330.

47.         The process of the delayed notification of the further appeal has delayed the preparation of this decision; it is essential that parties wishing to bring matters before the Tribunal have ensured well in advance of the hearing that appeals have been made in respect of every matter or period which they wish to be considered on appeal.

The Parties’ arguments

48.         At the beginning of the hearing we encouraged the parties to discuss the order in which they were to put their arguments, as the burden of proof in respect of the appeal against the initial assessment and the Additional Assessment fell on Serla Ltd, but the burden of proof in respect of the s 60 Penalty, the Sch 24 Penalty and the Additional Penalty fell on HMRC. After a brief adjournment for them to discuss the position, they informed us of their agreement that HMRC’s case should be put first, followed by the case for Serla Ltd.

Arguments for HMRC

49.         We deal below with Mr Hiller’s arguments on the facts, together with those of the Appellant. Mr Hillier referred to s 73(1) VATA 1994. The issue of “best judgment” in the context of VAT assessments had been considered by Woolf J in Van Boeckel v Customs and Excise Commissioners [1981] STC 290 at 292. On an appeal against a “best judgment” assessment, it was for the Appellant to displace the assessment, otherwise it must stand good.

50.         In relation to penalties, two different regimes applied because of the periods covered. Three periods were within s 60 VATA 1994. Under s 60(1) VATA 1994, the maximum penalty was 100 per cent of the tax. Mr Hillier referred to Ghandi Tandoori Restaurant v Customs and Excise Commissioners (1988) VAT Decision 3303, and R v Ghosh [1982] 2 All ER 689.

51.         The other periods were within Sch 24 FA 2007. Mr Hillier referred in detail to the relevant provisions; we consider these below in the context of the facts.

52.         HMRC submitted that the appeals should be dismissed and the assessments and penalty assessments confirmed, subject to the question of recalculation (considered below). Mr Hillier confirmed that HMRC did not intend to make further assessments.

Arguments for Serla Ltd

53.         Mr Osam’s arguments for Serla Ltd were principally on matters of fact, which we consider below. However, in relation to Van Boeckel he referred to the Badridas case cited by Woolf J at p 293. He submitted that the system being used in the present case did not separate the maker of an innocent mistake from an intended defaulter. Van Boeckel was silent in relation to individuals who were innocent mistake makers. We consider this argument later in the context of the factual evidence.

Discussion and conclusions

54.         In his skeleton Argument, Mr Osam submitted that this case was about best judgment, and that where human errors might have occurred, best judgment had been formed on irrelevant enquiries to the exclusion of relevant matters.

55.         Mr Hillier referred to the comments of Woolf J in Van Boeckel; we note that in subsequent cases, the courts have further considered the approach in “best judgment” appeals. In Customs and Excise Commissioners v Pegasus Birds Ltd [2004] STC 1509 (CA), Carnwath LJ referred to the statutory background to the expression:

“[10] The term 'best of their judgment' is derived from s 73(1) of the 1994 Act:

'Where a person has failed to make any returns required under 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.' (Emphasis added.)

It should be noted that the shorthand 'best judgment', as used in some of the cases, may be misleading, if it is taken to imply a higher standard than usual. The statutory words 'to the best of their judgment' are used in a context where the taxpayers' records may be incomplete, so that a fully informed assessment is unlikely to be possible. Thus the word 'best', rather than implying a higher than normal standard, is a recognition that the result may necessarily involve an element of guesswork. It means simply 'to the best of (their) judgment on the information available' (see Argosy Co Ltd v IRC [1971] 1 WLR 514 at 517 per Lord Donovan).”

56.         Later in his judgment, Carnwath LJ commented:

“[16] In Rahman (1), I drew attention to phrases used by Woolf J in the leading case under this Act (Van Boeckel v Customs and Excise Comrs [1981] STC 290) and in previous authorities in other tax contexts, to explain the effect of the 'best of their judgment' requirement (see [1998] STC 826 at 835):

'The passages I have italicised show that 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 has been reached “dishonestly or vindictively or capriciously”; or is a “spurious estimate or guess in which all elements of judgment are missing”; or is “wholly unreasonable”.”

57.         He explained at [22] that the part following the latter cited passage from Rahman (1) was a possible source of confusion, so we have omitted that part of the citation.

58.         At [72], Chadwick LJ commented:

“At the risk of stating the obvious, the power conferred by s 73(1) can be exercised only for the purposes for which it is given; that is to say it can be exercised only for the purposes of assessing the amount of VAT due to the best of the Commissioners' judgment. To purport to assess an amount of VAT due from a taxable person which is not the amount due to the best of the Commissioners' judgment is an improper exercise of the power. That is not in dispute. The issue raised by this appeal is whether, in the exercise of the s 73(1) power, more is required of the Commissioners than an honest and genuine attempt to make a reasoned assessment of the VAT payable, on the basis of the material then available to them. Is it enough, as the judge held, that the officer through whom the Commissioners act in making the assessment 'does his honest best'; or is there some objective standard against which the assessment must be measured so that, if the officer fails to attain that standard, there has been no proper exercise of the power to assess and the assessment must be treated as if it had not been made?”

59.         Chadwick LJ went on to express doubt as to the appropriateness of any objective standard. At [91] he continued:

“In para 44 of my judgment in Rahman (No 2) I suggested that, in cases where the tribunal had material before them from which they could see why the Commissioners made the assessment that they did and it was not apparent on the face of that material that the power to assess had not been exercised in accordance with the 'best of judgment' requirement, the tribunal would be well advised to concentrate on the question 'what amount of tax is properly due from the taxpayer?'; taking the material before them as a whole and applying their own judgment …'.”

60.         In Mithras (Wine Bars) Ltd v Revenue and Customs Commissioners [2010] UKUT 115 (TCC), [2010] STC 1370, Judge Oliver QC referred to the nature of the First-tier Tribunal’s jurisdiction:

“[11] The principles established in Van Boeckel and Rahman 1 indicate that the FTT's jurisdiction when considering whether an assessment was raised to the best of the commissioners' judgment is akin to a supervisory, judicial review type jurisdiction. The FTT does not have a true appellate function in that it cannot set aside the assessment on the basis that it disagrees with the commissioners' decision to make the assessment. The circumstances in which the FTT can decide that the assessment was not raised to the best of the commissioners' judgment, and therefore should not have been made at all, are very limited, essentially being restricted to cases where the commissioners have acted perversely or in bad faith. Carnwath J in Rahman 1 indicated that this 'kind of case is likely to be extremely rare' and that in the normal case 'it should be assumed that the Commissioners have made an honest and genuine attempt to reach a fair assessment': see page 836 of the judgment.”

61.         Later Judge Oliver QC made clear that the jurisdiction of the First-tier Tribunal when considering the amount of the assessment was not restricted in that way; the Tribunal’s function was truly appellate, in that it could consider further information or argument at the hearing of the appeal and reduce the amount of the assessment, thereby substituting its own view on quantum for that of HMRC.

62.         As Mr Hillier submitted, the burden of proof in relation to the VAT assessments falls on Serla Ltd to show that they should not stand; see Pegasus Birds at [14]. The position is different for the s 60 Penalty and the Sch 24 Penalty; under s 60(7) VATA 1994, the burden of proof lies on HMRC, and as accepted in argument by Mr Hillier, it also falls on HMRC where a penalty is imposed under Sch 24 FA 2007 for a deliberate inaccuracy where that inaccuracy is “deliberate and concealed” as defined in para 3(1)(c) Sch 24 FA 2007.

63.         We deal first with the assessments. Mr Hillier submitted that the issue was whether the VAT returns of Serla Ltd for the VAT periods 08/08 to 11/11 understated the tax due to HMRC as a result of not declaring all the business takings. We agree that this is a relevant part of what we have to determine, but on the basis of the issues raised by Mr Osam, we also have to consider whether the assessments were made to the best of HMRC’s judgment.

Mr Patel’s evidence

64.         Mr Patel gave the following evidence. At the initial unannounced visit on 7 October 2010, he carried out an investigation of the shop till, which he identified as a Casio TE 2200. He ascertained the original program settings, ie the preset till settings as programmed for use by the business. He printed out Z readings and X readings from the till. He identified that there were several program settings which required changing, and applied new system presets to the till; he obtained a till print stating that new presets had been defined. In cross-examination, he explained that the reason for making the changes was to make the cash register compliant, as his investigations had shown that it was not compliant when he had started his checks of the till. In response to the Tribunal’s question, he indicated that he had made the settings compliant with a “support sheet” containing HMRC’s technical guidance relating to the programming of tills.

65.         He explained that he attempted to interrogate the “clerk programme” settings; the sign-on code did not allow him to advance past clerk 2, and an alarm warning from the till was audible when he tried to sign on from clerk 3 to clerk 15.

66.         At the second unannounced visit on 21 October 2010, he obtained from the till a last receipt report, a Z reading and a ZZ1 reading. (As these are more generally referred to as Z2 readings, we use the latter term in this decision.) He also obtained the program settings on the till, and confirmed that the amendments which he had made to the presets on 7 October 2010 had not been changed.

67.         We accept Mr Patel’s evidence, which was not challenged by Mr Osam.

Ms Arridge’s evidence

68.         Ms Arridge gave evidence relating to the same two unannounced visits. She stated that at the first visit Mr Candemir was asked to count the money in the till. This amounted to £134.43. A further £35.00 in banknotes was found in a box adjacent to the till, and £129.00 in coins was found in the cupboard under the till. Mr Candemir confirmed that he had paid a supplier £229.87 cash from the till earlier that day.

69.         After giving details of the period, hours and payment for his employment, as well as cash payments to suppliers, Mr Candemir indicated that all employees’ wages were paid in cash from the till. He explained that if he needed to open the till for any reason other than a sale, he would ring up a sale with a £0.01 value; he did not use the “no sale” key. Cancelled sales and refunds were made as needed using the “-ve” key on the till.

70.         Ms Arridge had a telephone conversation with Mr Duzgun during the inspection. He explained that he normally cashed up the till between 11.00 pm and midnight and took a Z reading at that time; the last Z reading had been taken at around 1.20 am that morning before the shop had closed.

71.         When asked whether the Z2 report facility was used, Mr Duzgun initially said that he used it from time to time to check the till and that he had last run a Z2 report at about 18.00 the previous day. A few minutes later he retracted this statement and said that he did not use Z2 reports. Several times during the conversation Mr Duzgun said to Ms Arridge that he was feeling very panicked and did not know how to reply to her questions.

72.         During the visit, suspected counterfeit or non-duty paid alcohol and cigarettes were removed from the premises by the HMRC detection team. Other suspected counterfeit goods were removed by the Trading Standards officers. At the end of the visit, used journal till rolls and till reports were taken by the HMRC officers and a receipt issued to Mr Candemir.

73.         At the second unannounced visit in October 2010, the individual operating the till was Mr Toyam, who described himself as an employee. He was asked to count the cash in the till. He explained that of the £285 in banknotes, £100 belonged to someone else who had asked the business to look after it. £50 in a drawer outside the till was said to belong to Mr Duzgun personally and had been set aside to pay a bill.

74.         Mr Duzgun, who had been contacted by telephone at the start of the visit to ask for his agreement to allow the inspection to proceed, arrived approximately 30 minutes after the start of the inspection and was issued with an authority letter and a copy of HMRC’s factsheet CC/FS4. He confirmed that the £100 belonged to a third party and that he had left the £50 to settle a debt. He said that he used the X1 facility two or three times a day to check the cash in the till and that a Z report was used to work out the daily gross takings. Z reports were left in the shop until taken to his home every few days.

75.         He stated that he completed loose leaf daily takings sheets provided by his accountant to record the Z report figures and that in cases where the cash reconciliation figure differed from the Z report, the Z report was recorded as the correct sales figure. The takings sheets and the supporting till reports were given to the accountant along with all the purchase and wages records each quarter and that the VAT calculations and completion of the return were done by his accountant. Mr Duzgun said that he did not check his accountant’s calculations before signing the VAT return. At the end of the visit, till reports were taken up and a receipt was issued to Mr Duzgun.

76.         Although in cross-examination Mr Osam raised questions as to the approach taken in questioning Mr Duzgun, this did not amount to a challenge in respect of the points which we have recorded above. We accept this evidence given by Ms Arridge.

Ms Harter’s evidence

77.         In her evidence, Ms Harter explained that Serla’s VAT returns had been selected for assurance as part of a multi-agency exercise. The purpose of assurance visits was to ensure that the correct amount of VAT had been declared on traders’ VAT returns at the correct time. Her reference to “multi-agency” was to different sections of HMRC.

78.         Following discussions with Officer Nowak after the meeting at Zek & Co on 15 February 2011, Ms Harter had concluded that it was necessary to undertake another interrogation of the cash register. She and Officer Mandy Sleight, who interrogated electronic cash registers and interpreted the results which they produced, had made their unannounced visit to Serla’s premises on 1 June 2011.

79.         Ms Harter had tried to interrogate the till, but found that it was password protected. Officer Mandy Sleight then provided her with a password to access the till records, and Ms Harter extracted various reports from the till.

80.         One of the reports showed 15 clerks (ie identities against which till transactions could potentially be recorded); there were two clerks identified as C01. The first was at the beginning of the list, and the second appeared at the point which would have been occupied by C07. Officer Nowak had previously identified this latter clerk as a training clerk.

81.         Ms Harter referred to various other reports which she had printed during the visit; we consider these below.

82.         Ms Harter concluded that VAT had been understated by Serla Ltd because when she analysed reports extracted by herself and Mr Patel, she found that transactions with the clerk C01 belonging to the position normally occupied by clerk C07 did not appear on the journal roll and did not appear on the Z readings. Mr Duzgun had told her during the meeting on 15 February 2011 that he used the Z reading figure to complete the daily takings sheets which he gave to the accountants, and that they then used these figures to complete the VAT returns.

83.         In addition, she concluded that it was unlikely that there would be no trade for such long periods of time as indicated by the analysis; she attached an Appendix to her statement setting out these details; again, we refer to these below.

84.         Serla Ltd had admitted that there was an understatement of sales on 1 June 2011; this admission had been made in the accountants’ letter of 28 October 2011. Ms Harter had looked at the total of the two Z reports taken on that date, and concluded, by comparing the total with those in the sales day book for the period from 26 May 2008 and 30 November 2011, that there were only four days when the takings exceeded the 1 June 2011 total.

85.         The accountants for Serla Ltd had said that the inconsistencies between the customer receipts and the journal roll were due to a mechanical fault with the machine, that it was physically damaged and that the till had to be switched off and on. (We note that in the letter from Zek & Co dated 28 October 2011 they record Mr Duzgun as having said that he would switch it on and off again whenever it crashed, as all computers do.)

86.         Ms Harter had considered this explanation and concluded that this did not explain the understatements. Turning off and on the cash register would not clear the memory. She concluded that a more likely reason was that the training clerk was being used to suppress sales from the Z readings. The explanation which had been given did not elucidate how a password was required in order to alternate from the regular C01 clerk to the C01 training clerk.

87.         Ms Harter explained how the assessments in respect of the VAT and the penalties had been calculated. We consider this at a later point in this decision.

88.         We accept the above evidence from Ms Harter, which was not challenged by Mr Osam. (We refer later to the issue of the manner in which the visits to the business premises of Serla Ltd were conducted.)

Mr Nowak’s evidence

89.         In his evidence, Mr Nowak indicated that since mid-2005 he had specialised in the examination and interpretation of cash register operations and reports; he assisted other HMRC officers in relation to visits made to retail businesses to establish whether their VAT declarations were accurate. He had also co-authored internal guidance on the interpretation and use of cash register reports in the course of VAT audits.

90.         Mr Patel and Ms Harter had supplied him with a number of original documents, which he recognised as being electronic cash register reports and receipts from a Casio cash register. He had examined these and was satisfied that they showed that within the cash register there existed programming enabling the suppression of a proportion of sales rung through it as “training sales” yet being genuine sales.

91.         A “training clerk” was allocated on the printout of “Programme 3”, which showed the settings of a number of options available for a number of clerks on the cash register. Line 67 in position D6 counting from the right under the relevant clerk heading was a “1”.  The instruction manual for this model of till showed that this instructed the till to treat this as a training clerk. No other clerks were coded in a similar way.

92.         There were two clerks that appeared with a label “C01”. That appearing between clerks “C06” and “C08” was the training clerk, which in reality was clerk 7.

93.         The periodic Z2 reports were programmed to contain sales made by both normal clerks and training clerks; this was shown by the coding in another part of “Programme 3”. The instruction manual for the till showed that this instructed the till to include sales made by both normal clerks and training clerks in the Z2 report. This option was not available for the Z report usually taken as a daily report (ie the Z1 report). The latter was the one said by Serla Ltd to be used to arrive at the daily takings which were given to the accountant in order to produce the VAT returns.

94.         The value of sales made by a training clerk could be quantified by printing specific reports from the cash register.

95.         In the course of Mr Nowak’s evidence, Mr Osam made an application seeking disclosure of the support sheet forming part of the HMRC guidance and its inclusion in the evidence. We considered that application and concluded that it would not be appropriate for that guidance to be made public, and that in any event it would not assist in establishing the evidential matters necessary for the purposes of the appeal. We therefore refused the application.

96.         Mr Nowak exhibited a receipt showing a sale made by clerk C01. This showed the date 22 September 2010, and the “Check No.” as OOOEAA. The sale was for grocery valued at £1.79, the customer paying £1.80 and receiving 1p as change.

97.         He had compared this transaction with the portion of the journal roll where this transaction should be listed, the consecutive number of this transaction being 141211. He noted that on the journal a date was printed for each transaction, but no time was shown. Using the cash register’s programming, the journal time had been suppressed from printing.

98.         The consecutive number 141211 appeared on the journal, but the transaction recorded under that number was for a different amount (£0.59) and bore a different “Check No.”, OOOAAA. Reference to the clerk programming enabled identification of that Check No. as belonging to the genuine clerk C01 (the first entry in the list of clerks) and of Check No. OOEAAA as belonging to the second clerk C01, ie the one appearing in the place of clerk 7, and being a training clerk.

99.         Clerk 7 had had its default identification of C07 altered to C01 to give the appearance of sales made by this training clerk as having been sold by the genuine C01. A further programme amendment had been made from the default setting to inhibit the appearance of asterisks on a training clerk receipt; the “standard” demonstration example exhibited to Mr Nowak’s witness statement also showed the word “Trainee” at the position where the clerk number would appear for a normal sale. Mr Nowak expressed the view that concealed sales could be made with little fear of discovery unless the detail of the programming was obtained to show that the two clerks labelled as “C01” were differently programmed.

100.     Mr Nowak provided other examples, which we do not set out at this point, as their effect is similar. He referred to the contention by Serla Ltd that any errors due were due to a mechanical error, and pointed out that to enter the training mode and use the training clerk on this till, a four digit password was required; this was shown by the clerk programming details, the second line of the second “C01” including the digits “1980”. He considered it more likely than not that the code was entered to use clerk 7, rather than any mechanical error being involved; in his view, this, along with the manner in which the till had been programmed, showed evidence of concealment.

101.     Further evidence of concealment could be seen in a line of the programme relating to the printing of Z2 reports; this showed the password 1980 being required to input the Z2 report that would show the value of sales including those made by the training clerk.

102.     Mr Nowak made clear in his oral evidence that the changes made by Mr Patel did not affect the operation of the till, but simply instructed the till to print more information in order to be compliant for the purposes of the VAT information to be provided to HMRC. The support sheet did not contain any request to make changes to the clerk programming.

103.     Although detailed, Mr Osam’s cross-examination of Mr Nowak did not challenge Mr Nowak’s evidence so far as it related to matters of fact, and we therefore find as fact the factual matters stated in his evidence. To the extent necessary, we consider matters of opinion later in this decision.

Mr Zekai’s evidence

104.     Mr Zekai, the proprietor of Zek & Co Accountancy Services, explained the firm’s relationship with Serla Ltd. It was the firm’s practice to provide its clients with some form of formal recording process, by means of daily takings sheets. Serla Ltd had informed the firm that in the riots, the shop had been attacked and that the till and goods from the shelves had been taken away by the mob.

105.     The firm did not know how the till had been operated and used by Serla Ltd, nor how the takings were counted.

106.     We accept the above evidence.

Mr Duzgun’s evidence

107.     Mr Duzgun explained that he managed the business by depending on on-site employees, limiting his management to daily accounting for the float in the till, counting the daily takings at around midnight, authorising purchases, giving or allowing removal of funds if kept on the premises for purchases, and hiring and firing employees of Serla Ltd.

108.     He dealt with the documentary evidence used for his VAT returns. This meant taking Z readings and completing daily takings record sheets. He kept all the purchase invoices and other evidence of trade at the shop, as well as his regular Z readings.

109.     He indicated that he assisted his bookkeepers and accountants to the best of his and his employees’ ability and capacity, referring to the ability of everyone concerned to operate electronic and mechanical devices and manual back-up systems in sometimes difficult circumstances. He said that during the riots the shop had been looted and the cash register removed.

110.     In examination in chief, he stated that he did not know what “clerks”, as referred to in Mr Nowak’s evidence, were; he thought this was a name. He had not changed the program sequence, nor had he asked anyone else to do so. He had never previously seen a printout of “Programme 3” as exhibited to Mr Nowak’s statement [ie the list of clerks]. He did not know how there could have been a difference between the time shown on the print-out of the daily Z reading taken on 31 May 2011 (reading showing 23.03) and what was said to be the actual time, 00.44. He did not know how to change the time on the till.

111.     Although he remembered having had a telephone conversation with Ms Arridge, he could not remember the details. He denied having said what she had she had stated that she did about the times of his working in the shop and cashing up and taking a Z reading. He stated that he carried out those processes immediately after closing the shop. He also disagreed with Ms Harter’s statement that he had given the shop opening hours as opening between 07.00 and 08.00 and closing between 23.00 and 00.00 during the week and 01.00 at the weekend. There was no definite closing time; he checked the street each night and decided when to close.

112.     In order to open the till when there was no sale, he and Mr Candemir pressed the 1p key; there was no other way of doing so.

113.     His explanation for gaps in trading was that he was in control of many things, and for family and other reasons he had to close the shop to pick up his children or to obtain supplies or go to the bank for Paypoint matters.

114.     In relation to Mr Nowak’s evidence, Mr Duzgun stated that he did not understand the technical and mechanical issues concerning the cash register.

115.     Mr Duzgun’s evidence was tested in cross-examination by Mr Hillier. We consider the relevant issues below, and reach our conclusions based both on the points made in cross-examination and on the remainder of the evidence.

The issues in this appeal

116.     Mr Hillier submitted that the following issues needed to be considered:

(1)          Was there an understatement of VAT that needed to be assessed?

(2)          If so, was Serla Ltd liable to the s 60 Penalty?

(3)          Similarly, if so, was Serla Ltd liable to the Sch 24 Penalty and the Additional Penalty?

(a)   Best judgment

117.     We agree with Mr Hillier that these issues require to be considered, but we need first to consider whether there is any basis on which Serla Ltd can show, in the words of Chadwick LJ in Pegasus Birds, that the power to assess had not been exercised in accordance with the 'best of judgment' requirement.

118.     In doing so, we take account of the remainder of his comments as set out above. However, Mr Osam did seek to argue that HMRC had abused the goodwill shown by Mr Duzgun. Mr Osam based his argument on the absence of any mention to Serla Ltd of references to alterations to empirical evidence to support HMRC’s hypothesis until the appeal was already in progress and the parties had to serve their evidence. Mr Duzgun had not been given any opportunity to discuss any of the evidence “covertly obtained” by HMRC until the hearing. In particular, the important print-out, of “Programme 3”, had not been made available by HMRC until service of the evidence. Had this emerged and been explained at an earlier stage, Serla Ltd would have been in a position to assist intelligently in relation to HMRC’s enquiries. Mr Osam submitted that Mr Duzgun had showed co-operation by allowing interference with the till. Mr Patel’s evidence in cross-examination was that he had been acting under Mr Nowak’s instructions; Mr Nowak said that the alterations to the till were made in accordance with the requirements.

119.     In the context of Van Boeckel and the reference to the Badridas case, Serla Ltd submitted that the system being used did not separate the innocent mistake maker from the intended defaulter; Van Boeckel was silent in relation to individuals who were innocent mistake makers. Mr Duzgun was not saying that he had made a mistake; he was saying that he thought he made a mistake, even though he had known that there had been no real purchase at £2.84. (We explain below the relevance of this amount.) He now knew that it was part of a hypothetical exercise.

120.     Mr Osam emphasised that Mr Duzgun had shown goodwill by allowing the HMRC officers to carry out unannounced visits, by allowing the prearranged procedure involving adjustment of the till’s printing-out process, and by making payment of additional VAT because of his belief that £2.84 not recorded in the till’s electronic memory was in fact part of a day’s takings. Mr Duzgun had thought that this was a sale when in fact it was not.

121.     We have considered the issues raised by Mr Osam concerning the assessment process. Bearing in mind the comments of Carnwath LJ and Chadwick LJ in Pegasus Birds, a trader seeking to argue in respect of a “best judgment” case that the actions of HMRC have been such as to call the validity of the assessment into question is under a very heavy burden.  As Carnwath LJ had said in Rahman 1, we would need to be satisfied that

“. . . the assessment has been reached “dishonestly or vindictively or capriciously”; or is a “spurious estimate or guess in which all elements of judgment are missing”; or is “wholly unreasonable” . . .”

122.     We are unable to make such a finding in relation to the present case. Unannounced visits are part of the process through which HMRC seek to ensure compliance by traders with their VAT obligations. Mr Duzgun took the decision to allow the visits, which showed a degree of co-operation on his part. If he had decided against the visits, this might well have encouraged HMRC to carry out enquiries in other ways.

123.     HMRC needed to establish whether the VAT returns submitted by Serla Ltd were correct or (using the relevant words from s 73(1) VATA 1994) “incomplete or incorrect”. In order to do so, they needed to check whether the amounts derived from the Z readings and daily takings sheets prepared by Mr Duzgun for submission to the accountants could be verified by examination of the till records. On the basis of the evidence given by the HMRC officers, we find that there were sufficient grounds for HMRC to consider that the amounts shown in the VAT returns did not necessarily reflect the full information available as a result of the investigations carried out by the HMRC officers. We consider below the question of the methodology in the context of the amounts assessed, but we are satisfied that the process cannot be described as unreasonable or capricious.

(b)   Was there an understatement of VAT?

124.     We therefore return to Mr Hillier’s list of issues, and to his first question concerning understatement of VAT. We are satisfied that the Z readings were used to record the daily sales figures; Mr Duzgun confirmed this to Mr Hiller in cross-examination. As a result, if the Z readings do not fully reflect the sales made by the business, this gives rise to an understatement of VAT.

125.     The reports which Ms Harter extracted from the till at the visit on 1 June 2011 were the clerk report, department reports, programming report, Z report and Z2 report; in addition, she extracted the journal roll.

126.     From the clerk report, she noted that the first clerk C01 was a normal clerk programmed with a label 000AAA. This label was printed every time this clerk made a sale, and appeared on the receipt for sales made only by this clerk. The second clerk was programmed with a different label 000EAA. It was therefore possible to distinguish between the two clerks identified as C01 when comparing receipts. Mr Nowak had previously identified the second “C01” clerk as being a training clerk.

127.     During the visit she also printed sales receipt from the cash register’s memory. This could only be done for clerks C01 to 03, the second C01, and C010. There was nothing in the memory for any of the other clerks.

128.     The first copy receipt which she printed was dated 1 June 2011. The time on the receipt was 11.59. It was numbered 199540 and the “Check No.” was 000AAA. The sale was for grocery at £0.59, the customer paying £10.00 and receiving £9.41 change. Ms Harter compared this receipt with the journal roll; the latter showed the same consecutive number 199540, with the same details. (We note that the clerk number shown on the receipt was C01.)

129.     She then examined a customer receipt for clerk C01. The date of the transaction was 1 June 2011, and the time shown 00.57. The transaction number was shown as 199409, and the Check No. was 000EAA. The latter identified the transaction as being recorded on the second clerk C01. The sale was for grocery at £2.84; the customer paid £10.04 and received £7.20 as change. She compared this with the journal roll. This showed a transaction with the same consecutive number 199409; however, the details shown on the journal roll were different, as the time shown for the transaction was 05.33 and the grocery amount was £12.48.

130.     She took a Z report for 1 June 2011 (the sequential number being 1008). The amount shown was £303.73, and it was timed at 12.24 hours.

131.     She also took a periodic Z2 report dated 1 June 2011. The amount was £306.57. The difference in value between the two reports (ie this and the Z report) was £2.84. She concluded that this was the transaction numbered 199409 recorded via the second clerk C01 with the Check No. 000EAA.

132.     In addition she took periodic Z2 reports for transactions and for the department (grocery). The total value for these transactions was £306.57, which was the same as the periodic Z2 report mentioned in the previous paragraph.

133.     However, the periodic Z2 cashier report showed that in relation to C01 (ie the “normal” clerk 1) the value was £303.73 and that in relation to the second C01 (ie the training clerk) the value was £2.84. Ms Harter concluded that this was the amount which had been suppressed from the daily Z report referred to at [130] above. She further concluded that the cash register was programmed not to include the sales made by the second clerk C01 in the Z report. However, these sales were included in the Z2 report.

134.     Ms Harter also examined the Z reports and journal rolls taken from the premises of Serla Ltd by her and by Mr Patel. The journal roll showed all transactions made between Z reports except by training clerks. Each transaction had a unique sequential number for all clerks except training clerks.

135.     It followed that any time between any two sequential transactions on the journal roll was, according to the journal roll, a time when no sales were made. She carried out an analysis and produced a record of her findings showing the periods of time when (again, according to the journal roll) there were no genuine sales.

136.     She concluded that it was unlikely that a shop in such a busy location would have no customers for such long periods of time.

137.     Ms Harter pointed out in her evidence that the times printed on the various reports were different from the actual times when the reports had been extracted by HMRC officers.

138.     She concluded that after the daily Z reading was taken on 31 May 2011 at 23.03 (her estimate of the actual time being 00.44 on 1 June), the till was used to make a sale using the second (training) clerk C01 and the sale (transaction number 199409 at 00.57, for £2.84) was not recorded on the journal. Yet on 1 June 2011 at 05.33 a sale was recorded with transaction number 199409 for £12.48 by the first (“normal”) clerk C01. (In relation to the actual time of the Z reading, we find that she was in error in calculating that this was 00.44; as the Appendix to her witness statement shows in all cases that the actual time of the various transactions mentioned was 41 minutes later than the time shown by the till, the time and date of the Z reading must have been 23.44 on 31 May 2011.)

139.     In the light of her conclusions that the VAT due had been understated, Ms Harter calculated the understated VAT to be assessed by taking the amounts shown on the Z2 reports on 7 October 2010, 21 October 2010 and 1 June 2011. The Z2 reports showed total sales (including those recorded in respect of the second clerk C01, the training clerk) totalling £763.26.

140.     She calculated that these sales were recorded over a total of 19 hours, the total of the time between the shop opening and the respective times (ie the actual times) when the Z2 readings were taken. This was on the basis that Mr Duzgun had told her that the shop was open from between 7 and 8 am. On the basis of the calculation, the average VAT-inclusive sales per hour figure was £40.

141.     As Mr Duzgun had told her that the shop was open seven days a week, she calculated the total hours open each week as 7 x 17 = 119 hours per week. (We comment below on this part of her calculation.) This gave expected weekly sales of 119 x £40 = £4,760. For a 13 week VAT period this gave an expected sales figure of £61,880 inclusive of VAT.

142.     She calculated the standard rated percentage of sales in each of the declared VAT periods. She applied this to the difference between the declared sales (inclusive of VAT) and applied the VAT fraction for standard rated sales to arrive at the amount of underdeclared VAT.

143.     In her letter to Serla Ltd dated 15 February 2012 she gave information relating to her calculations and attached a schedule showing details of a proposed assessment in respect of underdeclared sales. The amount to be assessed was £14,617.00. The date of issue of this assessment was 26 March 2012.

144.     Although HMRC subsequently found on checking the assessment that there had been an error in Ms Harter’s calculations, this was an error in favour of Serla Ltd, in that it understated the additional liability to VAT in respect of the periods covered. We do not consider that this error calls into question the method by which the VAT had been calculated. (We have other points to raise on the calculations; we return to these later below.)

145.     The way in which the till had been set up to operate as a result of the programme settings raised questions concerning the accuracy of the daily Z readings, which were the primary records used to prepare the daily takings sheets from which the VAT returns were prepared by the accountants for Serla Ltd. The use of the second clerk C01 as a training clerk meant that transactions could be entered into the till without forming part of those records. HMRC obtained evidence of transactions carried out using this clerk, and established that such transactions appeared only on Z2 readings; these transactions did not show on the journal rolls or the Z readings.

146.     Mr Osam argued that the approach taken by HMRC involved deriving a hypothesis from one transaction, that involving the sale of groceries for £2.84 at 00.57 on 1 June 2011 (transaction number 199409 with Check No. 000EAA). He submitted that no evidence from a real sale of real goods purchased by real persons had been used to sustain HMRC’s hypothesis. They had caused tremendous confusion to Mr Duzgun with their concealed hypothesis. Serla Ltd had made 25 per cent restitution on the basis of a hypothetical purchase; Mr Osam submitted that there had been no purchase.

147.     In our view, Mr Osam’s argument is not sustainable on the evidence. The evidence shows that HMRC took copies of other transaction records showing the Check No. 000EAA apart from that on 1 June 2011. As stated above, Mr Nowak exhibited to his statement a copy of a print-out with that Check No. for a transaction dated 22 September 2010. We find that Mr Nowak’s conclusions as to the potential use of the training clerk in order to conceal transactions were entirely justified on the evidence.

148.     Mr Nowak referred to other print-outs showing transactions with this Check No. One was dated 7 October 2010. Again, the clerk label was “C01”. The consecutive number of the transaction was 143930. This record was printed out before the changes made by Mr Patel to the till programming at the visit on 7 October 2010. In contrast to this record, the journal record of sales showed the sale with this consecutive number as clerk C03 with a value of £0.00 timed at 10.43; Mr Nowak stated that it was self-evident that this was not the same transaction. We agree and accept his evidence on this point.

149.     The other example given by Mr Nowak was another with the same Check No. dated 21 October 2010, again made by the training clerk labelled C01. The sequential number was 146101, the sale being for grocery at £0.55. The cash tendered was £0.70, the change being £0.15. The journal roll entry for that sequential number was for clerk C01 (Check No. 000AAA) and the amount of that transaction was £2.73. We agree with and accept Mr Nowak’s conclusion that this demonstrated the absence of the training clerk sales from the journal roll, with the result that such sales were omitted from the Z report.

150.     Thus HMRC established that transactions had been recorded using the training clerk, and that such transactions did not find their way into the VAT records of Serla Ltd used to produce its VAT returns.

151.     In our view, it was not necessary for HMRC to observe the shop in order to establish whether the VAT liability of Serla Ltd had been understated. As part of the exercise of seeking to make an assessment of the understated tax to the best of HMRC’s judgment, we find that the omission of transactions from the Z reports gave sufficient grounds for them to start the process of seeking to calculate the amount of that understatement.

152.     As stated above, Mr Duzgun initially told Ms Arridge on 7 October 2010 that he used the Z2 facility from time to time, and that he had last run a Z2 report at about 18.00 on the previous day. Shortly afterwards he retracted this statement and said that he did not use Z2 reports. In her evidence, Ms Harter explained that a Z2 (or ZZ or ZZ1) report was a periodic report; it showed the takings recorded since the last time that a Z2 report was taken. (In other words, taking a Z2 report results in the Z2 readings re-setting to zero.) Ms Harter referred to a Z2 report taken by Mr Patel on 21 October 2010. This showed sales of £286.26. Ms Harter took a Z2 report on 1 June 2011. This showed sales of £306.57. In her view, this demonstrated that Serla Ltd had taken Z2 reports between these dates. If not, the value of sales on the Z2 report taken on 1 June 2011 should show all sales since the last Z2 report taken by Mr Patel on 21 October 2010; this would be far in excess of £306.57.

153.     We accept Ms Harter’s conclusion. The value of sales between the Z2 report taken on 21 October 2010 and that taken on 1 June 2011 would have been far more substantial than £306.57; we therefore find that Z2 reports were taken between those dates, and that Mr Duzgun’s statement made to Ms Arridge, that he did not take Z2 reports, cannot have been correct.

154.     In her letter to Zek & Co dated 15 February 2012 indicating that she proposed to raise an assessment of £14,617.00, Ms Harter referred to the Z report which she had taken on Wednesday 1 June 2011 at 12.24; the total takings up to that point in the day were £303.73. Serla Ltd had produced a further Z reading for that day showing takings of £364.25; a copy of that Z reading had not yet been supplied to her as previously agreed at the September 2011 meeting. The total takings for the day were £667.98. Ms Harter had examined the sales day book and had noted that according to the sales day book, during the period from 26 May 2008 until 30 November 2010 (approximately two and a half years) there had been four days when the takings had exceeded that figure, and three of these had been Sundays.

155.     We have referred above to the method used by Ms Harter to calculate the amount of the understatement. Her calculation of the average hourly rate of sales was based on a total of 19 hours over which the sales recorded in the three Z2 readings taken on 7 October 2010, 21 October 2010, and 1 June 2011 would have been made from the opening of the shop at 7 am to the actual times when the respective Z2 readings were taken. This produced an average of £40 sales per hour.

156.     We calculate the total times between 7 am and the taking of the Z2 readings as about 16.5 hours, allowing for the difference between the times as recorded by the till and the actual times at which the HMRC officers took those readings. This would produce an average sales rate of £46.25. As that would increase the amount of the understatement, the error is in favour of Serla Ltd; we see no reason to interfere with this element of HMRC’s calculation. The fact that the error is in favour of Serla Ltd does not in our view affect the conclusion as to the methodology of the respective assessments.

157.     Although Mr Osam referred in his argument to innocent mistake makers, we do not think that the evidence supports his contention that Serla Ltd, in the person of Mr Duzgun, had made an innocent mistake. As the issue of the state of mind of the trader is more relevant to the question of liability to penalties, we return to this issue under that heading.

158.     As noted above, another factor referred to by Mr Osam and Mr Duzgun at the hearing was that Mr Duzgun disagreed with Ms Harter’s statement that the shop opened between 7 and 8 am and closed between 23.00 and midnight during the week and at 1 am at the weekend. He explained that at night he checked the street and then closed the shop; he had never referred to a definite time.

159.     On the basis of the various copy print-out receipts included in the evidence, it appears to us that there were some occasions when the shop was open much further into the earlier hours of the morning, and others when no transactions were recorded beyond 23.15. At times, therefore, the shop was open longer than the average taken into account by HMRC, which would favour Serla Ltd, and at other times the use of an average midnight closure time might appear to show an element of disadvantage to Serla Ltd. We find that HMRC’s use of the average amounts, in the words of Chadwick LJ cited above, to “an honest and genuine attempt to make a reasoned assessment of the VAT payable, on the basis of the material then available to them”.

160.     Although we question some elements of Ms Harter’s calculations of the periods during shop opening hours when according to the till records no trading took place, we do not consider that this affects the calculation of the underdeclared VAT. The apparent gaps between trading were simply a factor leading to her view that a calculation needed to be made based on the actual Z2 readings taken by HMRC officers on the three occasions.

161.     The additional VAT liabilities covered by the Further Assessment appear to us to be purely the result of arithmetical errors made by Ms Harter in calculating the amount of the original assessment, those errors being in favour of Serla Ltd.

162.     Despite the “rough and ready” nature of some elements of the calculations, we find that the methodology adopted by HMRC in arriving at the amounts of the assessments was sound. No question was raised at the hearing on behalf of Serla Ltd in relation to the detail of the calculations, and we consider that HMRC made appropriate use of such information as was available to them. We do not consider that any valid criticism can be made of the manner in which HMRC’s visits to the business premises were conducted.

163.     We find that there was an understatement of the VAT due, that the original VAT assessment and the Further Assessment were made to the best of HMRC’s judgment, as described by Carnwath LJ in Pegasus Birds, and that on the evidence before us, Serla Ltd has not satisfied us on the balance of probabilities that those assessments were incorrect or inappropriate.

(c)    Validity of the assessments

164.     In her letter dated 15 November 2012 concerning the Further Assessment, Ms Harter stated:

“Further to my letter dated 15 February 2012, I am writing to advise you that the assessment has been mis-calculated and under states the amount of additional tax due from you. As a result I intend to raise a further assessment which is two years after the end of prescribed accounting period. I refer to Section 73(6)(b) of the VAT Act.”

165.     We do not think that this is a correct statement of the position under s 73(6) VATA 1994. This provides:

“(6) An assessment under subsection (1), (2) or (3) above of an amount of VAT due for any prescribed accounting period must be made within the time limits provided for in section 77 and shall not be made after the later of the following—

(a) 2 years after the end of the prescribed accounting period; or

(b) one year after evidence of facts, sufficient in the opinion of the Commissioners to justify the making of the assessment, comes to their knowledge,

but (subject to that section) where further such evidence comes to the Commissioners' knowledge after the making of an assessment under subsection (1), (2) or (3) above, another assessment may be made under that subsection, in addition to any earlier assessment.”

166.     Before looking at the Further Assessment, we consider the original VAT assessment issued on 26 March 2012. This covered VAT periods 08/08, 11/08, 02/09, 05/09, 08/09, 11/09, 02/10, 02/10, 05/10, 08/10, 08/10, 11/10, 02/11, 05/11, 08/11 and 11/11.

167.     The overall time limit in s 77 VATA 1994 is four years. As the assessment was issued on 26 March 2012, all the VAT periods covered by the assessment were within this overall time limit. However, it is then necessary to consider the further limitations imposed by s 73(6) VATA 1994. A number of these VAT periods ended more than two years before 26 March 2012. The only basis on which those earlier periods could be validly assessed was if the alternative condition in s 73(6)(b) could be fulfilled.

168.     The process of checking whether the VAT returns of Serla Ltd showed its full liability to VAT began with HMRC’s first visit to the shop premises on 7 October 2010. This was followed by a second visit on 21 October 2010. After this, Mr Nowak carried out an analysis of the till records. The next stage was the meeting at Zek and Co on 15 February 2011. After that meeting, Ms Harter discussed the position with Mr Nowak, and concluded that it was necessary to carry out another interrogation of the cash register; this was done at the unannounced visit on 1 June 2011. Thus HMRC did not have the necessary information on which to base their conclusions as to understatement of the VAT liability until 1 June 2011. As the original VAT assessment was made within one year of that date, we find that it was made within one year after evidence of facts sufficient in HMRC’s opinion to justify the making of an assessment came to their knowledge. It was therefore this part of the assessment process which required the use of s 73(6)(b) VATA 1994.

169.     The Further Assessment was issued on 22 November 2012. This therefore fell over one year after sufficient evidence came to HMRC’s knowledge. It would not have been open to HMRC to use s 73(6)(b) VATA 1994 at that stage. The only basis on which an assessment could be made was under s 73(6)(a) VATA 1994.

170.     The VAT periods covered by the Further Assessment were 11/10, 02/11, 05/11, 08/11 and 11/11. The earliest of these periods ended just under two years before the date of issue of the Further Assessment, so that the condition in s 73(6)(a) VATA 1994 was met and that assessment was validly made within the time limit. In her letter dated 15 November 2012 Ms Harter was incorrect when she referred to s 73(6)(b) VATA 1994 in the context of the Further Assessment.

171.     We are aware that HMRC’s standard internal advice in their manual VAEC6210 is that in certain cases involving arithmetical errors, an assessment should be withdrawn and a new assessment issued in preference to making a supplementary assessment. However, the effect of doing so in the present case would have been that a number of VAT periods would by then have been “out of time” for assessment, and a significantly greater amount of the underdeclared VAT would not have been assessable. As it was not possible for the Further Assessment to include any VAT periods before 11/10, the assessable amount for all the earlier periods had to remain unchanged despite the calculation error in favour of Serla Ltd.

(d)   Liability to the s 60 Penalty

172.     For HMRC, Mr Hillier accepted that the burden of proof of dishonesty fell on HMRC. HMRC contended that Serla Ltd had taken action for the purpose of evading VAT and that its conduct involved dishonesty. Mr Hillier referred to Ghandi Tandoori Restaurant v Customs and Excise Commissioners (1988 VAT Decision 3303), in which the Tribunal referred to the further mental element in addition to that of intending to evade tax, and said:

“In the majority of cases under this section the course of conduct adopted by the taxpayer will be such that the necessary mental element of dishonesty can be readily inferred.”

173.     Mr Hillier referred to sales made by the training clerk, which did not feed through to the Z readings used by Serla Ltd for the purpose of preparing its VAT returns. As Mr Nowak had stated in his evidence, a password was required to be entered into the cash register in order to use the training clerk. HMRC did not allege that Serla Ltd had necessarily programmed the cash register to omit sales from the Z readings, but it had been programmed to do so. He emphasised that HMRC had no evidence to show that Serla Ltd itself had programmed the till to do this. However, it was clear that the functionality of the till had been used in a certain way and that the use of that functionality did amount to dishonest conduct.

174.     The process was deliberate, in that it required positive action to utilise the training clerk and ring through sales to be recorded against it.

175.     There was further evidence of dishonest conduct in relation to the Z2 reports. These showed all sales made by each of the registered clerks on the till. The Z2 reports cleared the till, so that it could be clearly established that they had been used; there was a record of the times when Z2 reports were taken. Mr Duzgun had stated in evidence that he had no knowledge of Z2 reports and that such reports had not been taken. HMRC contended that such reports were run by the business and it was clear that they had been taken. Mr Hillier referred to the three Z2 reports taken by HMRC officers on the three visits. As Z reports reset the memory, each Z2 report should show all sales made by regular and training clerks since the previous Z2 report. If it were the case that Z2 reports had not been taken, one would expect much larger amounts to be shown on Z2 reports than was the case.

176.     Between the first and the second of these Z2 reports taken by the HMRC officers, only £286.26 of sales were shown. On the third, taken six months later, the total was £306.57. HMRC submitted that it was inherently unlikely that in those months the business made only those sales. Mr Osam’s figures supplied at a later stage showed transactions of between £300 and £600 a day. Mr Hillier argued that it was inherently unlikely that Z2 reports had not been taken in between; in his submission, it was more likely than not in the context of Z2 reports that either Mr Duzgun or Cem Duzgun (who according to his witness statement had at times undertaken management responsibilities) had extracted such reports from the till. (Mr Hillier acknowledged that Serla Ltd was not relying on that witness statement.)

177.     HMRC argued that Serla Ltd knew the correct level of sales, and that keeping transactions away from ordinary Z readings amounted to dishonest conduct when measured according to the ordinary standards of honest and reasonable people. The need for use of a password to use the training clerk involved positive action in order to obtain information from the till. For Serla Ltd it had been argued that the till had been dropped, and thus possibly damaged, and that the errors produced by the till might be mechanical errors because of this damage. HMRC submitted that it was unlikely that any mechanical issue was the cause of these errors and the printing of the Z2 reports.

178.     Mr Duzgun had stated in evidence that the business knew how to switch between clerks. Given the need for a password, HMRC submitted that deliberate action was more likely to have been the reason for what had happened, in particular the errors in relation to the till records. All these things amounted to dishonest conduct in submitting the VAT returns.

179.     We find it inherently improbable that errors in the till records were the result of mechanical problems with the till. In addition, we find that Z2 reports were taken by Serla Ltd, as clearly the amounts shown on the Z2 readings taken by HMRC would have been very much greater if no Z2 reports had been taken in between those occasions. We are satisfied that HMRC’s findings in respect of the three Z2 reports are indicative of the position at other times, and that the approach of Serla Ltd in relation to the recording of transactions has been consistent over the periods covered by the original VAT assessment (and the Further Assessment). We do not accept the submission that innocent mistake was involved.

180.     In relation to the s 60 Penalty, we find that the conduct of Serla Ltd involved dishonesty and thus that it is liable to the penalty, subject to the question of mitigation under s 70 VATA 1994.

181.     In her penalty assessment letter dated 20 June 2012 notifying Serla Ltd of the s 60 Penalty, Ms Harter stated in the context of that penalty that the tax evaded was £2,045. No reduction under s 70 VATA had been allowed for disclosure; however, a reduction of 20 per cent had been allowed for co-operation.

182.     We have reviewed the extent of co-operation with HMRC by Serla Ltd and its advisers, and see no reason to adjust the 20 per cent reduction allowed by HMRC. In the absence of disclosure, we agree with HMRC’s view that there is no basis for any reduction for disclosure. We do not consider that there are any other grounds for reduction of the s 60 Penalty, which we confirm in the sum of £1,635.

(e)    Liability to the Sch 24 Penalty and the Additional Penalty

183.     Mr Hillier referred to paragraphs 1 and 3 of Sch 24 FA 2007 (not set out in full in this decision). The conditions for a penalty were that a document provided to HMRC contained an inaccuracy amounting to or leading to an understatement of tax, and that the inaccuracy was careless or deliberate on the part of the person giving the document to HMRC. Under paragraph 3 an inaccuracy was deliberate and concealed if it was deliberate on the relevant person’s part and that person made arrangements to conceal it, for example by submitting false evidence in support of an inaccurate figure. Under paragraph 4, the penalty payable was 100 per cent of the potential lost revenue; paragraph 5 defined the latter as the additional amount of tax due as a result of correcting the inaccuracy.

184.     There was no definition in Sch 24 FA 2007 of “deliberate and concealed”. HMRC argued that this included submission of a return knowing it to be incorrect and having taken steps before or afterwards to conceal the inaccuracy.

185.     For much the same reasons as those given in relation to the s 60 Penalty, HMRC contended that the inaccuracy was deliberate and concealed. The sales had been recorded via the training clerk, and a password had been used; someone at Serla Ltd would know that sales recorded in this way did not feed through to the normal Z reading.

186.     If HMRC’s contention was right, it was also the case according to Mr Duzgun’s evidence that he counted the cash in the till. Thus, leaving aside the question of the Z2 readings, there would also be a discrepancy between the Z reading figure and the cash in the till. It would therefore be known that the sales figure was incorrect, and this would feed through to giving HMRC an incorrect VAT return.

187.     The evidence also established that Z2 readings had been taken by someone working for the business, using the required password. This meant that someone in the business had been taking Z2 readings and knew the correct figure of sales.

188.     Steps had been taken to conceal the inaccuracy by utilising the training clerk to hide the correct reading. Without the journal rolls and transaction numbers, an observer would not readily see that the training clerk figures did not go through to the Z readings.

189.     As shown by Mr Nowak, the coding in the till had been used to designate the training clerk, which would normally have been clerk C07, as C01. Although the Check No. was different, the appearance given by the till print-out was that all the sales had been made via clerk 1.

190.     All these points together amounted to evidence of deliberate and concealed behaviour on the part of Serla Ltd. The result of that behaviour was that incorrect returns had been submitted to HMRC.

191.     We have already set out our conclusions on the arguments raised on behalf of Serla Ltd concerning the till records. A further point made by Mr Osam was that at the first visit, the HMRC officers had carried out a prearranged procedure which involved adjusting the till’s printing process. We do not accept his suggestion that in some way this was designed to enable HMRC to obtain evidence to support their hypothesis. Mr Nowak made it clear in his evidence that the changes made to the till did not affect the way in which it operated.

192.     Mr Duzgun told us that he had purchased the business in 2008, and that the till was already in the shop at the time of the purchase. HMRC accepted in their argument that it was not possible to establish who had been responsible for setting up the till to operate in the manner described in the evidence. We do not think it necessary to identify that person, and we accept that it could have been someone without any connection to Serla Ltd. However, we are satisfied on the evidence that the till was used to record sales against the training clerk rather than the “normal” clerk C01, with the result that such sales did not appear on the Z readings used to complete the daily takings sheets forming the records from which Zek & Co prepared the VAT returns for Serla Ltd. We also take into account the fact that the takings were counted each night and compared with the amounts shown on the Z readings; this would have shown the discrepancies resulting from the recording of sales via the training clerk, as such sales did not show in the Z readings.

193.     We reject the argument that innocent mistake was involved, and find that the behaviour was deliberate and concealed, and thus that Serla Ltd is in principle liable to the Sch 24 Penalty and the Additional Penalty. Under para 4(2)(c) Sch 24 the penalty would amount to 100 per cent of the “potential lost revenue”; under para 5(1) this is the additional amount of tax payable as a result of correcting the inaccuracy.

194.     Mr Hillier referred to paragraphs 9 and 10 of Sch 24 FA 2007, which contained a structured regime for reductions for disclosure. HMRC submitted that in the present case the disclosure was prompted; the HMRC officers had carried out visits to the premises, and corresponded with Serla Ltd and its representatives concerning the Z readings.

195.     Paragraph 10 Sch 24 FA 2007 provided for a reduction of the penalty to take account of disclosure. In her first letter setting out the explanation of the various penalties (the attached penalty schedules being incorrectly headed “Wrongdoing Penalty under Schedule 41 Finance Act 2008”) Ms Harter had set out the reduction given; this was 30 per cent for giving HMRC access to records.

196.     Although Ms Harter used the incorrect statutory reference, we accept that the information set out in the explanation was in accordance with the correct legislation, Sch 24 FA 2007. We do not consider that her error affects the validity of the explanation, or of the Sch 24 Penalty. The information provided in relation to the Additional Penalty referred correctly to Sch 24 FA 2007.

197.     We consider that the reduction given for disclosure was at the appropriate level, and we see no reason to adjust it.

198.     Ms Harter also considered whether a special reduction was appropriate under par 11 Sch 24 FA 2007, and concluded that there were no special circumstances which would lead HMRC to make a further reduction in the penalty. We find that she gave proper consideration to this question, and see no reason for any such reduction.

199.     As the reduction for disclosure was 30 per cent, this applied to the difference between the maximum and minimum penalty percentages, 100 per cent and 50 per cent respectively. This gave a reduction of 15 per cent, ie (100-50) x 30 per cent. This was applied to the maximum penalty percentage of 100 per cent, giving a penalty percentage of 85 per cent.

200.     Paragraph 17 Sch 24 FA 2007 sets out the Tribunal’s powers on an appeal against a penalty. Serla Ltd did not appeal against the amount of the Sch 24 Penalty or of the Additional Penalty; it appealed against the decisions made by HMRC that these penalties were payable by it. Under para 17(1) Sch 24 FA 2007, we affirm HMRC’s decisions that Serla Ltd is liable to pay these penalties.

Result of the appeals

201.     The appeals of Serla Ltd against the original VAT assessment, the Further Assessment, the s 60 Penalty, the Sch 24 Penalty and the Additional Penalty are dismissed.

Right to apply for permission to appeal

202.     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.

 

JOHN CLARK

TRIBUNAL JUDGE

 

RELEASE DATE: 25 June 2014