Hajishoreh (t/a Pizza 1) & Ors v Revenue & Customs [2009] UKFTT 90 (TC)

FTT-Tax

Hajishoreh (t/a Pizza 1) & Ors v Revenue & Customs

[2009] UKFTT 90 (TC) · 2009-05-06

[1]

[2009] UKFTT 90 (TC)

TC00058 Appeal numbers MAN/99/0895 MAN/99/1119 MAN/00/0240 VALUE ADDED TAX — assessments — dishonest evasion — take away food outlets — test purchases, cashing up and analysis of Z-readings — conclusion that takings understated — assessments made to recover tax — one business unregistered — whether registration threshold exceeded — yes — whether assessments undermined by claims of errors, hoax calls and closure of nearby competitors — evidence of such factors tenuous — assessments upheld DISHONESTY PENALTY — whether dishonesty established — yes — whether mitigation allowed by respondents adequate — yes — appeal dismissed FIRST-TIER TRIBUNAL TAX M KARIMI HAJISHOREH (trading as Pizza 1) M KARIMI HAJISHOREH and K SAGHAFI (trading as Chichini's) Appellants - and - THE COMMISSIONERS FOR HER MAJESTY'S REVENUE AND CUSTOMS (VAT) Respondents TRIBUNAL: Judge Colin Bishopp Members: Carole Roberts, Christine Owen FCA Sitting in public in Manchester on 19, 20, 21 September 2006, 5 March 2007, 8, 9, 10, 11 September 2008 Taher Nawaz FCA for the Appellant James Puzey, counsel, instructed by the General Counsel and Solicitor to HM Revenue and Customs for the Respondents © CROWN COPYRIGHT 2009 DECISION Introduction The appellants, Mujtaba Karimi Hajishoreh (known as Mr Karimi) and Kamal Saghafi, carry, or carried, on business selling take-away pizzas, with a delivery service for those customers requiring it. At the times material to these appeals, Mr Karimi was the sole proprietor of a shop at Kirkburton, near Huddersfield, from which he traded as "Pizza 1", while he and Mr Saghafi traded in partnership from another shop at Netherton, also near Huddersfield, as "Chichini's". It was directed that, for convenience, their appeals should be heard together. They are: By Mr Karimi alone against an assessment dated 2 September 1999 for VAT of £28,708 (an associated civil penalty imposed on him on 16 February 2000 for alleged dishonest evasion, amounting to £18,660, was withdrawn because of an error in notification); By both appellants against an assessment dated 12 November 1999 for VAT of £89,882, amended on 21 February 2003 to £87,342; and By both appellants against a civil penalty imposed on them on 16 February 2000 for alleged dishonest evasion, amounting to £58,423, reduced on 21 February 2003 to £33,930.

[2]The assessments were already about seven years old when the appeals first came on for hearing, and the periods with which we are concerned go back much further, to 1990 in the partnership's case. Three days were allotted for the hearing, an estimate which proved wholly inadequate—in the event, eight full days, plus another for the discussion of some procedural matters, were required and even then we had to receive the appellants' closing submissions, subsequently, in writing. Almost two years went by between the beginning of the hearing and its conclusion, largely because Mr Karimi was suffering from illness. While we attribute no blame for the fact that we are required to determine the appeals long after the relevant events—by the time the hearing began, more than six years had elapsed since the investigation took place—it is obvious that delays of this kind are very much to be avoided. The delay in this case led not only to the fading of the witnesses' recollections, but also to the fading of some of the documentary evidence to the extent that in parts it had become illegible.

[3]Some of the relevant facts were agreed or, at least, undisputed. Mr Karimi began trading as Pizza 1, initially in partnership with Mr Saghafi and another man, but they ceased to be partners in the business and Mr Karimi ran it as a sole proprietor from a date which is now not entirely clear but was before 1996. He sold the business in 1999. The business was at no time registered for VAT. Chichini's, which was at all material times a partnership between the two appellants, registered for VAT on 13 August 1990. We were told that, before then, there had been four partners owning between them four shops in the Huddersfield area, all known as Chichini's, but that the partnership dissolved, Mr Karimi and Mr Saghafi taking the Netherton shop, and the other partners the remaining three shops. Because of the proximity of his home to Pizza 1, Mr Saghafi continued to work there on most days of the week, and allowed others, particularly the staff, to believe he was still a partner in the business, while Mr Karimi worked mainly at Chichini's, which happened to be nearer his home. The Commissioners do not contend that Mr Saghafi was in truth a partner in Pizza 1, and nothing turns on the pretence.

[4]That Customs and Excise, as they then were, were interested in the two businesses became apparent to the appellants on 12 February 1999, when officers paid unannounced visits to the two shops as they were on the point of closing for the day. Their purpose was to observe the cashing up and to secure such records as were available. There was a good deal of dispute, to which we shall return, about the manner in which the visits were conducted, and on the detail of what occurred, but it is not disputed that the officers did observe the cashing up, noted the amount of the takings, and collected the limited documentary evidence which was held on the premises, including the "Z-readings" produced by the tills. Further records were produced by the appellants in the course of the next few days. What they did not know at that stage was that on previous occasions officers had visited the premises to make test purchases, aiming to ensure that they made the first and last purchases of the day in order to determine, from the sequential numbering of the receipts, how many sales had been made on each such day.

[5]On 26 February 1999 the appellants were interviewed, separately, and each made some admissions that the true takings had not been recorded. They were asked to undertake self-invigilation, that is to make (independently of the till) notes of the day's sales, and on 13 March 1999 Customs and Excise officers attended each of the shops in order to conduct invigilation of their own. That exercise, too, was the subject of dispute, to which we shall return. There followed some correspondence between the appellants' then accountant, a Mr Mistry, and the Commissioners, in which Mr Mistry made further admissions, to the effect that there had been understatement of the partnership's takings, but in which he maintained, despite the earlier admission, that the level of the takings at Pizza 1 was never so great that Mr Karimi should have registered for VAT. There was a further interview of the two appellants, together, on 23 June 1999.

[6]The Commissioners considered that they then had enough information to make the disputed assessments and to impose the penalties, and they duly did so. The penalty, in each case, was fixed at 65% of the tax said to have been evaded since the Commissioners accepted that the appellants had cooperated with them to some extent and that the maximum penalty of 100% of the tax evaded should be mitigated accordingly. Later it was realised that some mistakes had been made. The penalty imposed on Mr Karimi alone was withdrawn, as we have mentioned. The assessment against the partnership contained an error in that one of the periods which the Commissioners had intended to assess had been wrongly identified. The elimination of that mistake led to the reduction in the assessment we have already recorded. The duplication of one period, 05/91, in the notification of the penalty imposed on the partnership and the consequent error of identifying the penalty intended for each period with the immediately preceding period (for example the penalty which was intended to be imposed for period 08/92 was erroneously shown as the penalty for period 05/92) meant that a large part of the aggregate penalty had to be withdrawn, resulting in the other reduction we have mentioned. The facts

[7]We heard evidence from, in all, ten officers—Andrew Lumb, Jayne Wood, Neil Priestman, Angela McCalmon, Quentin Lewis, Sylvia Jones, Martin Mummery, John Kinghorn, Deborah Patterson and Thomas Simmons, and had the undisputed statement of one other, Deborah Richardson. We heard the oral evidence of both of the appellants. All of the witnesses had made written statements, most of the officers in 2000. That is fortunate since by the time they gave oral evidence they had live recollections of so little that, save for the assessing officers (Mr Simmons and Mr Mummery for Pizza 1 and Chichini's respectively), none could add anything of substance to their statements and the contents of their notebooks, copies of which were also produced to us. We do not intend hereafter to recite the evidence of the witnesses, one after the other, but instead to deal with the facts in chronological order, mentioning the identity of the witnesses only when it is necessary to do so. Many of the events we describe are relevant to both establishments, and we shall deal with them together when appropriate.

[8]The first event relevant to the appeals occurred on 25 June 1992 when Mrs Richardson undertook a routine visit to Chichini's. It was not disputed that she discovered that the appellants were not retaining their till rolls, or that she told them that they must do so in the future, though Mr Karimi—who, as both appellants agreed, dealt with the record-keeping of both businesses—told us he could not remember the meeting. He also said he did not recall receiving from Mrs Richardson a letter (a copy of which was provided to us) written shortly after her visit, repeating the instruction, but said that if he had, he would have handed it on to his accountant. It was not disputed that the appellants continued to discard their till rolls.

[9]The test purchases were made on various dates in February 1999, including the day on which the cashing up was observed. (Some test purchases had, it seems, been made in 1998 but their results were not used in the making of the assessments and we treat them as irrelevant.) Typically the officers placed an order either very early or very late in the evening, in the hope of being respectively the first and last customers of the day (the shops opened from approximately 5 pm to midnight). The tills produced receipts bearing sequential numbers, which the officers asked be given to them. It was easy to determine from a comparison of the first and last receipts the number of transactions which had occurred on each day, although the till at Pizza 1 was set incorrectly: the date it recorded on receipts and other printed material it produced, such as Z-readings, lagged behind the correct date by three months, while the recorded time was about 45 minutes in advance of the correct time. This fact led to some confusion, but we are satisfied that it was properly resolved.

[10]At about 11.45 pm on Friday 12 February 1999, two officers called at each of the shops, Mr Simmons and Mr Kinghorn at Pizza 1 and Mr Mummery and Mr Lewis at Chichini's. Mr Saghafi was present at Pizza 1 and Mr Karimi at Chichini's. Mr Simmons and Mr Mummery spoke to them, while Mr Kinghorn and Mr Lewis asked questions of the staff, and then the officers at each shop watched as Mr Saghafi and Mr Karimi cashed up, that is counted the cash in the till and took a Z-reading of the day's takings.

[11]There was comparatively little conflict between the parties about events at Pizza 1. The difference between the cash in the till, adjusted for the float placed in it at the beginning of the day's trade and for payments made from it, and the Z-reading (which was £375.20) was insignificant, and easily accounted for by small errors. A Z-reading for the previous day, showing sales of £265.40, was also found in the till. Mr Saghafi is recorded to have said that the weekly takings at Pizza 1 were between £1,200 and £1,300 per week, or a little more, that the takings at Chichini's were between £1,600 and £1,800, and that the latter had increased recently. In his evidence, however, he said that he had little real knowledge of the takings at either shop, since Mr Karimi dealt with all financial matters, and moreover that despite his pretence of being a partner, he had no interest in Pizza 1 and was not able to speak with any authority about the business. He also pointed out that English was not his first language; while it was adequate for dealing with pizza sales, he was unable to use it to discuss finance. In addition, because of longstanding health problems, he had some difficulties of recollection.

[12]The visit to Chichini's was much more contentious. It is common ground that there was a significant discrepancy between the cash found in the till and the takings shown by the Z-reading which Mr Karimi took from the till at Mr Mummery's request, that only part of the till roll for the day was found—the roll had been replaced during the evening—and that Mr Mummery and Mr Lewis carried out an extensive search, including emptying the rubbish bags. They said they did so in an attempt, unsuccessful as it turned out, to find the missing till roll and any other evidence of the day's takings. Mr Karimi contended that the officers had in fact been successful, and that they had thereafter refused to admit that they had that material, or to allow the appellants to see it. We shall return to this point. Mr Karimi complained too that Mr Mummery was very aggressive, that he raised his voice several times, that the search went too far in that the rubbish bags were emptied in food preparation areas, and that Mr Mummery appeared to have arrived with the intention of finding the evidence to support the conclusion he had already reached, namely that there had been suppression of the takings at the shop. Mr Mummery and Mr Lewis both said that Mr Karimi had agreed to their carrying out a search, but told us they were unable to recall much of the detail. They did not accept that the search had been carried out in an unacceptable manner, but did agree that they had been anxious to find the missing till roll if they could. Mr Lewis also told us that he had called at the shop early in the evening to make a test purchase and had overheard a member of staff comment that the till roll then in the till had almost run out.

[13]Here, the difference between the Z-reading and the cash found in the till was significant. The cash amounted to £346.44 (after allowing for the float and the wages of £60 to shop staff and of £30 to the delivery driver which Mr Karimi said that he had paid during the course of the evening); the sales for the day recorded by the Z-reading were £583.10. Even the total of the sales recorded on the partial till roll which was available exceeded the cash in the till. The discovery of the substantial difference led to another area of contention: Mr Karimi complained that Mr Mummery was unwilling to listen to his explanation of it, but instead simply took it as further evidence supporting the conclusion he had already reached.

[14]We have little doubt, from his comportment in the witness box and from what he said, that Mr Mummery was firm in his approach and that the search was thorough. It would, also, be unremarkable if Mr Karimi, surprised as he was by an unexpected visit from Customs and Excise officers, was very anxious, and he may well have thought that Mr Mummery was not paying attention to what he said. Examination of Mr Mummery's notebook, made up immediately after the visit ended, however, shows that he has recorded in some detail all of the comments which Mr Karimi contends he made, or tried to make, but to which, he said, Mr Mummery was unwilling to listen. We will deal with the substance of those comments later; at this stage we record only that we see no reason to think that Mr Mummery and Mr Lewis conducted the visit otherwise than properly. We should, however, mention that Mr Mummery did not ask at the time of the visit whether any money (apart from that used to pay wages) had been taken out of the till during the evening; we agree, as indeed Mr Mummery accepted, that it would have been an appropriate question to ask.

[15]Having realised that there was little written evidence of sales at Chichini's Mr Mummery arranged with Mr Karimi (whose position was that Mr Mummery insisted) that he would collect such other material as Mr Karimi could provide later that morning (it was by now about 1 am on Saturday 13 February) and Mr Mummery duly called for the purpose at Mr Karimi's home. He was able to take away for examination such records as the daily gross takings book, which included notes of the wages paid to shop staff and delivery drivers, and the purchase records. No till records (in respect of either shop) were provided then, but till rolls covering the subsequent days' trade were kept by the appellants, at Mr Mummery's request, and later handed to him.

[16]On 26 February 1999 Mr Karimi and Mr Saghafi were interviewed, separately, at the Commissioners' office in Bradford. The interviews were both conducted by Mrs Patterson and Mr Mummery; the appellants were each accompanied by Sandra McLean, an accountant. It is apparent from the transcripts of the interviews with which we were provided (we also listened to the tape recordings) that Ms McLean took little active part in the interviews, though it is not clear to us how well informed she was about the appellants' affairs at that stage. She had not, we understand, represented them before, but was standing in for Mr Mistry who was on holiday, and the interviews took place barely a fortnight after the appellants became aware of the investigation. Much of the relatively short time occupied by the interviews (approximately 45 minutes each) was taken up with an explanation by Mrs Patterson of the system by which Customs and Excise then dealt with suspected cases of dishonest evasion (she was responsible for that aspect of the investigation) and the provision by Mr Karimi and Mr Saghafi of an explanation of the manner in which the two businesses were managed, and nothing now turns on those parts of the interviews.

[17]The Commissioners rely, however, on Mr Karimi's admission that he had understated the takings of Pizza 1 believing, he said, that it would be necessary to register for VAT if the takings in any one week exceeded the registration threshold. Later, he conceded that he had realised that Pizza 1 should be registered but had understated the takings in order to make it appear that they were below the threshold. The conversation then turned to Chichini's. Mr Karimi repeatedly made the point that the takings varied considerably, sometimes because of the closure of competing establishments and sometimes simply because a lot of customers came for no identifiable reason, and he also emphasised the number of mistakes which, he said, the shop employees made when operating the till. He did not, however, concede any understatement of the takings at that shop. Mr Saghafi made no admissions of understatement of takings, though he repeated his impression that the takings at Chichini's were about £1,600 per week He did not go so far as the £1,800 he is recorded to have said at the visit on 12 February, and he offered explanations of the variations in the level of takings which were consistent with those which had been given by Mr Karimi.

[18]The appellants were asked to undertake self-invigilation, as we have related, and did so at Pizza 1 from 11 March to 23 March, and at Chichini's from 11 March to 2 April. The Commissioners did not consider that the records produced by the appellants in this exercise were reliable, and discarded them when making the assessments. On 13 March, a Saturday, there was simultaneous invigilation by a sequence of Customs and Excise officers at each shop. This exercise, too, was the subject of some dispute. The appellants' case is that at times, particularly at Chichini's, the volume of orders from customers calling at the shop and from those placing orders by telephone for delivery or later collection exceeded the capacity of the staff to prepare and cook the food, and rather than tell customers they could not meet an order they resorted to taking the telephone off the hook. The volume of trade at Chichini's on that day was high, they said, and in accordance with their usual practice the staff took the telephone off the hook for part of the evening. The invigilating officers took the view that this course had been adopted, not because the staff were unable to cope, but as a device to suppress the takings on a day when Customs and Excise could see for themselves exactly what the turnover was, and the officers insisted on its being replaced, to the extent that on at least one occasion an officer himself replaced the telephone.

[19]It is apparent from examination of the invigilation records produced on 13 March that the number of transactions recorded by the appellants and the officers respectively are close; the small differences are easily accounted for by simple errors. What is also apparent is that the number of transactions recorded at each establishment on that day is higher than the number recorded on any other day, including the remaining Saturdays for which details were available. At Pizza 1, Customs and Excise recorded 67 transactions; on the following Saturday the appellants recorded only 47. At Chichini's, the officers recorded 97; on the following two Saturdays the appellants recorded 76 and 83. The differences between these figures are large. To what extent, if at all, they are due to the appellants' having been prevented by the officers from taking the telephone off the hook, and to what extent they are due to understatement by the appellants, when not watched, of the true number of transactions is impossible to say. Whatever the cause, it is in our view clear that the officers were justified in their conclusion that the self-invigilation records could not be regarded as a reliable guide to the true volume of trade.

[20]We observe at this point that the takings on 13 March at Chichini's, at £564.50, closely matched those recorded on the Z-reading taken there on 12 February, at £583.10; similarly, the Z-reading at Pizza 1 on 12 February was £375.20 and the sales on 13 March amounted to £388.50. The appellants' own records indicate that the volume of trade on Fridays was very similar to that on Saturdays. These factors, in our view, support the Commissioners' contention that the Z-readings obtained on 12 February do accurately record the true takings of each shop for that day. Moreover, the takings recorded on those days for which the respondents have Z-readings exceed, by a large margin, those recorded on all other days in the appellants' daily gross takings records. We are not persuaded that the officers' insistence on the replacement of the telephone on the day of the invigilation has distorted the true takings of that day.

[21]On 14 May 1999 Mr Mistry wrote to the Commissioners. He stated that in about 1997, when a nearby pizza takeaway shop closed, the takings at Chichini's had increased by about £200 per week but the appellants had concealed the increase, basing their declarations on the turnover achieved before the closure of that shop. He also said that the takings had increased further following the closure of a nearby fish and chip shop, that closure having occurred only a few days before the visit on 12 February. The letter went on to confirm the substance of what Mr Karimi had said about Pizza 1 at his interview, that is that he had understated the takings from time to time but Mr Mistry maintained—contrary to what Mr Karimi had conceded at the interview—that the true annual takings had at all times remained below the registration threshold. No mention was made in this letter of a Chinese takeaway restaurant very close to Chichini's.

[22]That restaurant was, however, mentioned in the second interview, on 23 June 1999. The interviewing officers were, on this occasion, Mrs Patterson and Mr Simmons, and Mr Mistry attended to assist the appellants. The admission that £200 of the weekly takings at Chichini's had not been declared, made by Mr Mistry in his letter of 14 May, was mentioned and neither appellant sought to withdraw or modify it. The conversation then turned to the unannounced visit on 12 February. The appellants were told that Mr Mummery had concluded from what he discovered then that the understatement of takings was rather greater than £200 per week, and asked for explanations. Mr Karimi mentioned that the Chinese restaurant had been closed as the proprietors were taking an extended holiday, and repeated what Mr Mistry had said about the fish and chip shop. He added that 12 February was close to St Valentine's day, a fact which may have increased the takings. Staff mistakes were again mentioned as a significant distorting factor. Copies of the calculations which Mr Mummery had prepared were given to the appellants, and Mr Mistry promised to write to the officers with his comments once he had had the opportunity of considering them properly.

[23]Mr Mistry's response came in a letter of 27 July. He rejected Mr Mummery's then calculation of the understated takings at Chichini's, at £91,584, as "an insult", so much so that, he said, the appellants did not intend to cooperate further in the investigation. He did, however, go on to suggest a compromise upon the basis that the takings at Chichini's had been understated by an average of £500 per week for two years, a total understatement of £52,000 of which £7,744.68 represented VAT. He emphasised that he was making the suggestion without the appellants' instructions; we were told that the appellants did not accept that there had been understatement of that magnitude (or, even, at all), and were asked to disregard what was, in effect, a without prejudice offer. We do disregard it; a proposal put in these terms has very little evidential value. Mr Mistry's letter concluded by repeating the contention which had already been made that, despite the understatement of the takings at Pizza 1, the turnover had never exceeded the registration threshold. He added that the business had been sold. The assessments and the appellants' attack on them

[24]An assessment was sent to Mr Karimi, in respect of Pizza 1, on 28 August 1999 but only a few days later, on 2 September, it was replaced by another, we understand because an arithmetical error had been detected. The letter sending the replacement assessment was written by Mrs Patterson, but it was Mr Simmons who had made the calculations which led to it. He had concluded that there had been understatement of the takings not only in early 1999, but ever since the business was taken over by Mr Karimi alone in 1995, and he projected his calculations backwards in order to cover the whole of that period. In doing so he also determined that the turnover exceeded the registration threshold in about July 1996, with the consequence that the business should have been registered from 1 September 1996, and that VAT (amounting in all to £28,708) was due for the entire period from that date until 27 June 1999, which he took from Mr Mistry's letter to have been the date on which the business was sold. We were told at the hearing that while the business had been sold in June 1999, it had closed in March, soon after the officers' invigilation. A civil evasion penalty was later imposed but, as we have mentioned, was subsequently withdrawn.

[25]Mr Simmons started with Mr Karimi's record of his daily gross takings. Though he considered that the figures understated the true takings, he thought they would be a reasonable guide to the proportion which each day's sales bore to the weekly total. He then took what he considered to be the only reliable guides to the true takings, that is the Z-readings for 11 and 12 February which had been obtained at the unannounced visit, and the result of the officers' invigilation on 13 March, together with the information he had from the test purchases about the number of sales to be expected each day, and the average value of a transaction which he was able to calculate from the till rolls he had. Save to the extent we have mentioned, he rejected the recorded daily gross takings figures and the result of the self-invigilation as unreliable. By combining the information he had he was able to make the calculations which led to the assessment. As his arithmetical approach was not challenged, we shall not set out any further detail in respect of this assessment; Mr Mummery's similar approach and his calculations in relation to Chichini's are described below.

[26]Mr Mummery used the evidence of the takings obtained on 12 February 1999, the till rolls later produced for the trade from 13 to 21 February, and the takings of the evening on which Customs and Excise officers invigilated, 13 March. The Z-reading for 12 February showed sales of £580.10, after eliminating the purchase made by an officer at the beginning of the evening, which he also compared with the average sales declared (that is, recorded in the daily gross takings book, on which the VAT returns were based) for the preceding Fridays from 18 December 1998 to 5 February 1999, of £271.71. He too, though rejecting the daily gross takings record as an accurate statement of the true takings, took it as a reasonable guide to the relative volumes of trade on each day of the week. Analysis of the declarations showed that Friday's takings represented 20.04% of an average week's takings; using that figure it was possible to calculate that if £580.10 represented the true typical Friday takings, a typical week's takings amounted to £2,894.71. The average declared in the period selected by Mr Mummery (excluding Christmas week, which he thought would be unrepresentative) amounted to £1,355.55, a difference of £1,539.16. That difference suggested that the takings at Chichini's had been understated by about 53%.

[27]Mr Mummery was also able to work out the average transaction value from the till rolls which were produced later. He arrived at a figure of £5.15. He then calculated the takings to be expected on the days on which officers had made test purchases using the same method as that adopted by Mr Simmons in respect of Pizza 1, and, by again using the proportions, drawn from the appellants' own records, which each day's takings bore to those of a whole week, found that, by this method, the takings to be expected in an average week amounted to £2,632.77. If that was correct, the average declaration (£1,355.55) represented an understatement of 48.51%. The assessment he made, covering the periods from 11/90 to 11/98, the last before the period in which the initial investigations took place, was based on the assumption that the takings had been consistently understated by the lower of the two figures he had determined, slightly rounded down. Again, the arithmetical accuracy of his calculation was not challenged.

[28]In addition, Mr Mummery undertook an examination of the records of Chichini's purchases which were given to him by Mr Karimi. They were not in good order, but he was able to compare some of them with information he or other officers obtained from the appellants' suppliers. He came to the conclusion that the purchase records were incomplete in that the appellants had concealed some of their purchases in order, he thought, that the apparent value of their purchases should be consistent with their understated sales. Mr Nawaz, the chartered accountant who represented the appellants at the hearing, was able to demonstrate to our satisfaction that in this part of his analysis Mr Mummery had made a number of mistakes, by missing some invoices which were in fact in the appellants' records, and of arithmetic. Mr Mummery agreed that there were some errors, and overall their extent is such that in our view an assessment based on an understatement of purchases would be difficult to uphold. However, though Mr Mummery considered that the omissions he believed he had detected from the purchase records supported his other calculations, he did not base the assessment he made on them. Mr Nawaz sought to make a great deal of the errors but, significant though they were, they do not, in our view, undermine Mr Mummery's assessment, provided the calculations and assumptions on which it is in fact based are sound.

[29]Mr Simmons also looked into the purchases at Pizza 1, but confined his analysis to purchases of the boxes in which pizzas were placed once cooked. However, he came to the conclusion that this analysis was of little help and he did not use it in making the assessment. We agree that the available information is uninformative, and Mr Nawaz's criticism of Mr Simmons' analysis of it, even if justified, does not advance the matter.

[30]Mr Nawaz's remaining arguments, which he pursued with some determination, were that the assessments did not take account of errors, changes of mind by customers, hoax purchases or, in respect of Chichini's, the closure of other outlets shortly before the investigations took place, the last being a factor which significantly increased the volume of the appellants' trade. A further criticism, that the calculation of the average transaction value ignored "no sales" and was correspondingly too high, can be easily dismissed since it quickly became apparent from an examination of the calculations that the criticism was misplaced, and Mr Nawaz did not pursue it further.

[31]The errors, as both the appellants told us as they gave evidence, were made in the use of the tills by young and inexperienced counter staff; Mr Karimi and Mr Saghafi, we were told, concerned themselves with the preparation and cooking of food, and left the handling of cash to other staff. The mistakes usually consisted of the incorrect keying into the till of a customer's order; similarly a customer's order may have been keyed in before the customer changed his mind, and asked for something different. The facility for correcting errors was, we were told, limited: an error could be corrected before the transaction was finalised but, once the order had been fully entered, it was not possible to make a retrospective correction. The result was that the till rolls showed sales which had not in fact taken place and, in consequence, the Z-readings taken at the end of the trading day showed totals which included the value of the non-existent sales.

[32]While we do not doubt that errors occurred—it would be remarkable if they did not—we are not persuaded that they are a material factor.

[33]It became apparent that the tills used at the two shops are of a sophisticated type, programmed in order that a customer's order is keyed in by reference to the goods ordered rather than the value of each item. The system employed by the appellants was that a customer's order, whether taken over the counter or by telephone, was entered into the till by pressing buttons to which each item sold by the appellants had been individually allocated. The till produced a "ticket" showing each item the customer had ordered, with a price per item, which was used in order that the food preparation staff knew what to cook. The same information was reproduced on the till roll. As the customer placed his order, the items were keyed in. When the order was complete an appropriate button was pressed and the till calculated and displayed the value of the order, for which the customer paid. In the case of an order placed by a customer present in the shop, as the officers' evidence showed, payment was requested immediately the order was complete and the food was then prepared; those placing telephone orders paid when the food was collected, if they called for it, or on delivery if delivery had been arranged.

[34]While it is easy to understand how one might press the wrong button by mistake when keying in an order, it seemed to us that the mistake would usually be immediately apparent from the electronic display. What we find difficult to understand is how some of the mistakes described by the appellants might have occurred. We were provided with copies of some till rolls on which the appellants had marked what they believed to be mistakes. It is sufficient to take three examples for present purposes. An order placed at Pizza 1 for a single item, identified as "1/2 cheese", at £3.10 is marked as a mistake. The immediately following recorded sale is identical. Both orders are shown by the till roll as having been paid for. We did not understand from the appellants' evidence how a duplication such as we are asked to accept could occur; it seems to us that it would require the conscious keying in of a single order twice. Later, on the same roll, a recorded purchase of another single item, "½ chilli" is marked as a mistake; the immediately following item is "¼ chilli". Here, it is easier to accept that a mistake might have been made, that the customer pointed it out when asked to pay more than he expected, and that the correct order was then entered.

[35]The third example we were given was one mentioned during the course of the interviews. It was suggested that the large discrepancy between the Z-reading and the cash found in the till at Chichini's on 12 February might be accounted for by the keying in of "£120" when what was intended was "£1.20", or possibly by keying in an order for three pizzas of a particular kind as one for thirty pizzas. The former suggestion seems to us impossible, since the till, as it was described to us, allowed the keying in of a money amount only to record the customer's payment, from which the till calculated the change which should be given. We could find no instance of the manual keying in of the price of an item since, as we have indicated, every item sold was pre-allocated. The incorrect keying in of the amount tendered would not affect the sales total for the day. The suggestion that 30, rather than three, pizzas might have been entered came from Mr Nawaz rather than the appellants. In our view it is so unlikely that such a large error went unnoticed—and Mr Nawaz did not identify it on the part of the till roll which was available for the day—that the possibility can be disregarded.

[36]On the Pizza 1 till roll obtained on 12 February an entry has been marked "change mind", indicating that the customer changed his mind about the order after it had been keyed in. It was a fairly large order, consisting of six items with an aggregate value of £15.90. The immediately following recorded sale has seven items, to an aggregate value of £16.80. The obstacle in the appellants' way in persuading us that there was a change of mind is that not one of the items in the first order appears in the second. While we can accept that a customer might change his mind about one or two items after placing an order, it is difficult to believe that an entire order might be changed. We find this claim implausible.

[37]The appellants also told us that they were the victims of hoax orders. Purported customers rang up to place an order, giving a false name and address, and when the delivery driver arrived with the food he found that the address did not exist, or the occupier of a genuine address denied any knowledge of the order. The sale had already been entered in the till but the price was not received, and the effect of hoax calls was to inflate the apparent takings, as recorded by the Z-reading. The till rolls produced to us are marked with what the appellants claim were hoax orders. Some, they said, were placed by rival establishments with a view to harming the appellants' trade, others by members of the public who derived malicious pleasure from the practice. Mr Mummery told us that he observed, at Chichini's, a note of two addresses to which the appellants would not deliver, and we accept that hoax calls are not pure invention. For reasons we shall develop later we are not, however, persuaded that they were as frequent as the appellants claimed.

[38]The respondents accept that some at least of the establishments which the appellants claim were closed during the relevant period were in fact closed. However, they contend that with one possible exception the effect on the appellants' trade would have been slight, at best. The fish and chip shop was in another village, about 2½ miles from Chichini's, and closer to another fish and chip shop which remained open. The pizza take away shop which had been identified was also about 2½ miles from Chichini's, and Mr Mummery discovered that it had closed some years before, and that there were other pizza outlets nearby. This evidence was not challenged, and we agree that the effect on the trade at Chichini's of these closures would have been negligible. The Chinese take away was, however, within a few doors of Chichini's and if it was closed there is evident merit in the appellants' claim that customers intending to call at that restaurant and finding it closed would instead have made a purchase at Chichini's. The respondents did not challenge, by positive evidence, the appellants' claim that the Chinese restaurant was closed on 12 February (the officers told us they accept that there was such a restaurant but did not remember whether or not it was open) though they point out that the evidence about the period for which it was closed was very vague, and that its closure was not mentioned until much later. We will return to the manner in which we think this factor should be taken into account shortly. Conclusions

[39]We begin by reminding ourselves of the approach the tribunal is required to adopt. The Privy Council put it succinctly in Bi-Flex Caribbean TA \s "Bi-Flex Caribbean" Ltd v Board of Inland Revenue (1990) 63 TC 515:

"The element of guess-work and the almost unavoidable inaccuracy in a properly made best of judgment assessment, as the cases have established, do not serve to displace the validity of the assessments, which are prima facie right and remain right until the taxpayer shows that they are wrong and also shows positively what corrections should be made in order to make the assessments right or more nearly right. It is also relevant, when considering the sufficiency of evidence to displace an assessment, to remember that the facts are peculiarly within the knowledge of the taxpayer."

[40]As that passage and many other authorities to the same effect show, the burden of satisfying us that the assessments are wrong rests on the appellants, and that remains the case despite the imposition of a dishonesty penalty: see Khan v Customs and Excise Commissioners

[2006] STC 1167

and, by analogy, Morris and another v HMRC (2007) 79 TC 184.

[41]Mr Nawaz complained, vigorously, that he had been hampered by what he perceived to be inadequate disclosure by the respondents, a topic which had been the subject of a number of interlocutory applications. Even by the time of the hearing, he said, disclosure by the respondents remained incomplete to the extent that it was impossible for the appellants to be sure of having a fair hearing. It is certainly right, particularly when a dishonesty penalty has been imposed, that HMRC should be open and forthcoming about disclosure, going voluntarily beyond the bare requirement of rule 20(1) of the Value Added Tax Tribunals Rules 1986 that a party must serve a list of those documents "which he proposes to produce at the hearing". (The replacement of this rule, in rule 27(2) of the Tribunals Procedure (First-tier Tribunal) (Tax Chamber) Rules 2009, requires little more.) We accept that there is something in Mr Nawaz's complaint that he has encountered some difficulty in securing from the respondents all the documents for which he has asked.

[42]However, Mr Nawaz's argument seems to us to miss the more important point that, had the appellants kept proper records in the first place, there would have been no need for him to demand the disclosure of whose inadequacy he now complains, and it ignores too the point made by the Privy Council in the extract we have quoted that the facts are within an appellant's own knowledge. The position might be different if this were a case in which the Commissioners have made assessments based on material which they have themselves created, such as observation evidence, and which they have only selectively disclosed to the appellants. But both these assessments are based on the—fairly limited—material which was obtained from the appellants, essentially a comparison of the declared takings with those identifiable from the unannounced visit on 12 February 1999, the subsequently produced Z-readings and the invigilation.

[43]In addition, much of his complaint, and Mr Karimi's allegation that Mr Mummery had secured some till records on 12 February 1999 which he then concealed, were contrary to Mr Karimi's own evidence. He told us that when a mistake was made the erroneous till ticket was placed in the till, in order that the error could be identified and taken into account when the cash in the till was reconciled with the Z-reading. Only then were the tickets discarded. It is clear that the reconciliation of the cash and the Z-reading had not taken place by the time Mr Mummery arrived, and accordingly any tickets preserved as evidence of errors should not by then have been discarded, yet Mr Karimi did not suggest that any such tickets were in the till. Mr Karimi was quite unable (or unwilling) to tell us what had become of the till roll which had been in the Chichini's till at the beginning of the evening's trade. He did not say that it had been thrown away (so that Mr Mummery might have recovered it); had he done so the obvious question would have been why it had been discarded before the end of the day's trade, and before it could be used as an aid to the reconciliation of the cash in the till and the Z-reading. We have concluded that Mr Mummery did not find, nor later conceal, any such records because there were none to be found.

[44]It is in our view difficult if not impossible to understand why, if they are honestly declaring their takings, traders who, like the appellants, have invested in sophisticated tills do not take care to ensure that they are used properly and correctly, that the information they produce in the form of Z-readings is accurate, and that the information is preserved. Instead, we are asked to accept that errors (which, according to the appellants, always had the effect of increasing the recorded takings and never the reverse) were commonplace and that the few Z-readings which were available when the unannounced visit was made on 12 February 1999 are unreliable. If traders fail to keep, or choose to conceal, reliable evidence of their turnover they can in our view scarcely complain that they are unable in consequence to mount an effective challenge to an assessment. It is true that some of the corroborative evidence used by Mr Mummery, that relating to the purchases, was not initially available to the appellants, and that Mr Mummery made some mistakes in his analysis of it, but it is equally fair to say, again, that had the appellants kept properly organised records there would have been less potential for the confusion which in fact arose. In short, we are not persuaded that any documentary difficulties under which the appellants laboured were not of their own making. We reject Mr Nawaz's argument that any disadvantage the appellants may have suffered in consequence of their own failings should be considered to have deprived them of a fair hearing.

[45]The requirement imposed on a taxpayer attempting to challenge an assessment, as the Privy Council made clear in Bi-Flex Caribbean TA \s "Bi-Flex Caribbean" , is to produce evidence to support a different figure. Mr Nawaz produced alternative calculations but, leaving aside for the moment the closure of the Chinese restaurant, they are based not on credible evidence but on mere assertion. While we are willing to accept, as we have indicated, that mistakes were made, and we are also willing to accept that occasionally the appellants were the victims of hoaxes, we are not satisfied from the evidence we heard that mistakes were uncorrected and that they invariably resulted in an exaggeration of the true takings, and we are not satisfied that hoaxes were of such a frequency that they had any significant impact on the level of takings. For reasons on which we shall expand later, we did not consider that the appellants were truthful witnesses and we do not accept their evidence about the frequency and effect of mistakes, or the frequency of hoaxes. The claim that St Valentine's day, on 14 February, might inflate sales of take-away pizzas on 12 February seems to us to be fanciful and indicative of an attempt to find any reason, however implausible, to explain away what the appellants know to be inaccurate records.

[46]By contrast, the evidence secured by the respondents is compelling. There was, as we have said, a close match between the takings and the Z-reading at Pizza 1 on 12 February. That fact, albeit a single comparison, undermines the assertion that the mistakes made by the staff when using the till were uncorrected and led to a significant difference between the true takings and the apparent takings recorded in the Z-reading. It is significant too that the takings achieved on the day of the officers' invigilation, at both shops, were comparable with those determined on the occasion of the unannounced visit, and that the takings recorded on those days, as the appellants were compelled to concede, exceeded by a considerable margin those recorded for any other day in the daily gross takings records on which the VAT returns were based. The exercise of multiplying the number of transactions determined from the test purchases by the calculated average transaction value led to a close correspondence, and we are satisfied from all this evidence that the respondents' calculations of the takings of the two businesses in February and March 1999 are as accurate as can reasonably be expected.

[47]The questions remain, however, whether the takings at Chichini's at that time can be taken as a fair guide to the takings in earlier periods, and may be relied upon to make an assessment covering, as in this case, a period of nine years, and whether they should be adjusted to take account of the distortive effect of the closure of the adjacent Chinese restaurant. At first sight it seems appropriate to make an allowance, but we are not satisfied that we should do so, and for two reasons. First, it is incumbent on the appellants to produce evidence of the effect the closure of that shop had, sufficient to enable us to make a fair estimate of that effect. We have no such evidence, and are being asked to speculate. As in case of mistakes, we are asked to assume that the effect was always to increase the apparent level of the takings over their true level. The fact that the appellants sought to claim this effect as the consequence of the closure of other outlets too far away to have any discernible impact on their takings, as well as the unexplained failure to mention the closure of this restaurant until much later, make it impossible for us to treat the claim as credible.

[48]Second, and perhaps more importantly still, the appellants' own records and the remaining reliable evidence do not bear out their argument that its closure did make a significant difference to their takings. The daily gross takings record of Chichini's provided to us, which covers the period from the end of December 1997 to 12 February 1999, shows steady levels of takings, increasing slowly over the entire period from about £1,250 per week to about £1,350 per week. There are very few weeks when the takings were significantly higher. Occasionally the takings were lower, but in those weeks it is apparent that the shop was closed, for example at Christmas. What is equally clear is that the takings recorded on a daily basis in the few weeks before 12 February show no significant change at all. In addition, as we have commented, the takings on 12 February and on 13 March 1999 (by which date the Chinese restaurant may have re-opened—the evidence on this point, despite its claimed importance to the appellants' case, was vague and quite inconclusive) were comparable.

[49]In the light of all those considerations, we have come to the conclusion that it is impossible for us to come to a view about the impact, if any, of the closure of the Chinese restaurant on the turnover at Chichini's. Any allowance we made would not even be a guess, but an arbitrary adjustment to take account of no more than a possibility that there was some impact. It is not open to us to make an adjustment on such grounds; as the authorities make clear, we may adjust an assessment only if we have evidence before us which "shows positively what corrections should be made". For those reasons we decline to adjust the assessments for alleged mistakes, changes of mind, hoaxes or the closure of other establishments.

[50]Less controversially, we were invited by Mr Nawaz to increase the allowance of about 2% which is made in the assessments for the sales by the appellants of zero-rated goods, consisting of salads and similar items. He was not ambitious, asking us to increase the allowance only to 3%. But the reason he gave was that "the appellants feel that sales in summer of salads are higher than those in winter months when the officers looked at the aspect of zero rating". They may be right but, again, we are being asked to make an arbitrary adjustment to allow for a mere possibility. For the reasons we have already given, we are unable to do so.

[51]The attack on the extrapolation backwards of the true takings calculated in early 1999 was not challenged on arithmetical grounds, but by reference to the other factors we have already dealt with and have found not to be established. In the absence of reliable evidence that there were material differences in the past, the respondents' approach to the calculation of past arrears seem to us to be unassailable. Accordingly, subject to one further point with which we deal below, we must dismiss the appeals against the assessments. Dishonesty

[52]Though he did not concede any understatement of the takings at Chichini's, Mr Nawaz did not advance a positive case that the appellants had not been dishonest; nor, in our view, could he realistically have done so. Mr Karimi admitted that he had understated the takings at Pizza 1 in order to conceal the fact that the registration threshold had been exceeded. His understanding of the registration rules may have been incorrect, but his intention was nevertheless the dishonest one of avoiding the registration for which he suspected (and, indeed, admitted at the first interview, knew) he should have applied. Understatement at Chichini's of £200 per week was admitted at the June interview. The appellants sought to retract those admissions when they gave evidence. We have found that the true takings at both establishments were much greater than the appellants maintain. We have already mentioned that, despite having acquired sophisticated tills, the appellants did not retain the information produced by those tills.

[53]The obvious and in our view compelling explanation is that they did not do so because that information would have revealed that their declared takings were incorrect. We are satisfied that Mrs Richardson's letter was received, and that the appellants knew perfectly well that they should have retained their till rolls. The claim that their English was poor and that they did not fully understand what was required of them is not borne out by the tape recordings of their interviews or by the manner in which they gave evidence. There does not seem to us to be any possible room for doubt that their conduct was dishonest and that they intended by that dishonesty to evade payment for the tax for which they knew they were liable. We are satisfied that the appellants set out to put forward explanations for the detected understatements of their turnover which they knew to have no foundation in truth, that the limited admissions they made were designed to deflect the respondents and, later, us from the true scale of their under-declarations and that they can not be regarded as reliable witnesses.

[54]Mr Nawaz made the point that there is (or, at the time, was) an ordinary limit of three years for the making of assessments of this kind, imposed by s 77(1) of the Value Added Tax Act 1994, and he invited us to discharge the assessments so far as they related to periods outside those three years. However, sub-s (4) of that section extends the period to 20 years "if VAT has been lost … as a result of conduct falling within section 60(1) …". Section 60(1), now repealed for most purposes but in force at the time with which we are concerned, was engaged when, "for the purpose of evading VAT, a person does any act … and his conduct involves dishonesty …". It follows from the conclusions we have set out above that the subsection is engaged, and that the relevant time limit has not been breached. We shall not, therefore, discharge any part of the assessments on these grounds. We add, for the avoidance of doubt although Mr Nawaz did not argue the contrary, that we are satisfied that the effective date of registration of Pizza 1 was calculated correctly.

[55]We come last to the question of mitigation of the remaining penalty. The respondents, as we have mentioned, allowed 35% by applying their usual tariff, which permits reductions from the maximum 100% of the tax evaded by reference to the extent and quality of the cooperation afforded by the taxpayer. We are not bound by the tariff, though it is usual, in the interests of consistency, for the tribunal to respect it, and we may increase or decrease the extent of the mitigation allowed as we think fit in the light of the evidence available to us. Mr Nawaz argued that the appellants had done everything asked of them, by attending interviews, undertaking self-invigilation and allowing invigilation by officers, and that 35% was too little. Mr Puzey, counsel for the respondents, argued that the amount allowed was sufficient since such cooperation as the appellants had given was superficial, that they had continued to attempt to conceal the true value of their turnover, and that they had forced the respondents to prove their case at a lengthy hearing, while continuing to deny the extent of their understatement.

[56]We agree with Mr Puzey and with his reasons. We are satisfied, as we have indicated, that the appellants pursued before us spurious arguments which they knew to have no foundation. In our view the mitigation allowed by the respondents was on the generous side, but it is not a case in which we think any part of the mitigation allowed should be cancelled. We are, however, satisfied that there is no material before us on which we could properly increase it.

[57]All three appeals are therefore dismissed. We direct that the appellants pay the Commissioners' costs, to be the subject of detailed assessment by a costs judge of the High Court if they cannot be agreed. COLIN BISHOPP TRIBUNAL JUDGE RELEASE DATE: 6 May 2009