VALUE ADDED TAX — Fast food takeaway — assessments for two shops based partly on records provided by appellants and partly on assumptions made by Customs — assessments for third shop based solely on increased declaration of takings by appellants — no issue on best judgment — all assessments reduced to nil — appeal allowed
CIVIL EVASION PENALTY — penalty falls with tax assessments — appeal allowed
MANCHESTER TRIBUNAL CENTRE
MR HOSSAIN ALEAGHA AND MR JACK MIRZIZAD
T/A PIZZA CARINO Appellant
- and -
THE COMMISSIONERS OF CUSTOMS AND EXCISE Respondents
Tribunal: Mr J D Demack(Chairman)
Mr A E Brown FCA
Mr J P Denny
Sitting in public in Manchester on the 3, 4 and 5 March and 23 and 24 April 2003
Mr Richard Barlow of counsel instructed by AVC Leeds for the Appellants
Mr Nigel Poole of counsel instructed by the Solicitor for the Customs and Excise for the Respondents
© CROWN COPYRIGHT 2003
DECISION
Introduction
- This is an appeal by Mr Hossain Aleagha and Mr Jack Mirzizad, trading as Pizza Carino, against:
1) assessments to tax of £91455 (plus interest) for the period between 20 December 1986 and 30 June 1998 inclusive (the assessment period) notified on 9 April 1999; and
2) a civil evasion penalty assessment of £91441 for the assessment period notified on 17 June 2000, which penalty was mitigated by 5 per cent to £86869 for the appellants' co-operation in determining their true liability to tax. (The co-operation consisted in their attending a single interview).
- The appellants appealed against the tax assessments because "the assessment has not been made by way of best judgment and is excessive", and against the penalty assessment because "suppression is denied".
- Parol evidence was given by:
Messrs G A Lumb, Q Lewis, S Gill, J Kinghorn, M D Mummery, T A Simmons, Mrs J Wood, Miss L C Bower and Ms J Hardwick, all of whom are Customs' officers;
Messrs S B and N G Tranmer, former suppliers of burgers to the appellants;
Mr Abdullah Patel, bookkeeper for the appellants; and the two appellants themselves.
- That evidence was supplemented by witness statements of Customs officers Messrs J Ingle and A G Clough, and Mesdames H T Walsh and S Illingworth, to which no objection had been raised, and a bundle of copy documents from each party. We labelled the appellants' bundle, P1, and the Commissioners, C1.
The facts
- We find the following facts to have been established.
- Messrs Aleagha and Mirzizad carry on the business of a fast food takeaway in partnership under the style of Pizza Carino. They sell pizzas, burgers, French fries, garlic bread, etc. Throughout the assessment period they traded from premises at 8A Old Bank Road, Mirfield, West Yorkshire. From 1991 or 1992 they also traded from 39 High Street, Heckmondwike. And for about 10 months from the end of 1997 they also traded from Seymour Arcade, 29 Towngate, Ossett. (The Appellants continue to trade from Mirfield. They sold the Heckmondwike business about 2 years ago, and closed the Ossett shop in October 1998). For 18 months to 2 years whilst hoping to obtain planning permission to extend the fast food shop at Mirfield the appellants also ran an adjoining card shop but as their planning application was unsuccessful they sold the card shop. The appellants explained, and we accept, that they closed the Ossett shop because it proved less profitable than they anticipated, the car parking facilities available being inconvenient for customers.
- Each shop was open daily from 5 pm to midnight, except that on Fridays and Saturdays Heckmondwike closed at 12.30 a.m. And from an unknown date not long after its opening, Ossett opened at 4.30 p.m. The shops took orders by telephone and operated a delivery service, that service being free at Ossett from a date in the spring of 1998 for orders above a specified value.
- Each shop had an electronic till programmed so that every single item on the menu appeared on its key pad. A cash customer would place his order on entering the shop. The cashier would key it into the till causing the till to open. Two copies of the till receipt would be produced, one for the kitchen as an instruction to the chef, and the other for the customer if he wished to take it. The customer would then pay for the food ordered, and receive change as appropriate. Payment could be made in cash or by cheque, but not by credit card. Telephone orders would be keyed into the till in the same way as cash customers' orders. Some telephone orders were initially noted on pieces of paper before keying in. The cashier's copy of the till receipt for telephone orders would be endorsed with the customer's name and be retained by the cashier. On the customer calling to collect and pay for a telephone order, the cashier would take his money and open the till by pressing the "no sale" key. If an order were to be delivered, the delivery man might have a cash float to enable him to give change as necessary, in which case he would account for his takings at the end of the evening. The takings would then be entered in the till by the cashier using the "no sale" key. Alternatively, the delivery man might take change for a £20 note on delivering individual orders, in which case his obtaining of change and accounting for the money for the order would generate two "no sale" keyings into the till. The till receipt for each order to be delivered indicated that it was a delivery.
- In evidence, Mr Lumb, the Commissioners' assessing officer, accepted that the practice of using the "no sale" key described above was by no means peculiar to the appellants and their business. He also accepted that use of the "no sale" key was necessary on any occasion the appellants wished to open the till, notwithstanding no true sale transaction was being dealt with. He further accepted that the evidence of "no sale" entries on the till could rise to a level of 50 per cent at times. We treat those acceptances as facts.
- We were not told what stimulated Customs' interest in the appellants, but between 2 March 1998 and 8 March 1998 officers visited the Mirfield and Heckmondwike shops, making a series of "first and last" purchases. To count the number of transactions recorded on the tills each day a purchase was made shortly after each shop opened and another just before it closed. The purchasing officer obtained a numbered till receipt for his or her transaction. By that means Customs were able to count the number of "transactions" carried out during each period of trading. Such "transactions" included "no sales" keyed into the tills. We find that nothing observed by the officers between 2 March 1998 and 8 March 1998 indicated that the appellants were suppressing takings.
- Based on what he described as the Commissioners' general experience of fast food establishments, Mr Lumb then applied a figure of £5 to the estimated number of sale transactions to produce an approximate takings figure. That exercise indicated to him that the appellants might be underdeclaring their liability to VAT. Consequently, he recommended that there be further investigation of their affairs.
- On Saturday 10 October 1998, Customs officers attended all three of the appellants' shops as they were about to close.
- Officers Simmons and Lewis attended Heckmondwike, where Mr Aleagha was working. A Z reading was taken in their presence in the sum of £709.30. The officers noted that cash and cheques in the till totalled £676.10, including a float of £120. We find that the difference between the £709.30 and £676.10 was accounted for by delivery drivers' floats. (In evidence, Mr Lumb relied on the fact that he had been told by Mr Simmons that the cash and cheques counted had totalled less than the Z reading to indicate the suppression of takings by the appellants. We are satisfied that Mr Simmons confused the evening's takings with the takings plus the floats so that, including the deliverymen's floats, we find the cash counted matched the Z reading). The officers were unable to obtain a till audit roll. Officers Kinghorn and Hardwick visited Ossett. They obtained a till audit roll for the whole of the week then ending and a Z reading for the day. The cash and cheques for the day matched the Z reading. Officers Lumb and Gill went to Mirfield where Mr Mirzizad was working. They obtained a till audit roll for the 6 days then ending and the Z reading for 10 October. Cash and cheques matched the Z reading. We observe that in no instance did the cash and cheques counted exceed the till reading, which we would have expected had there been suppression of takings. We find that the officers' activities on 10 October 1998 revealed no evidence of suppression of takings. (Indeed, as the procession of Customs' witnesses as to events on 10 October lengthened it merely served to strengthen the appellants' case.)
- In the October 1998 period for which the Commissioners obtained till rolls and Z readings, those documents revealed that at Ossett "no sales" varied between 27.7 per cent and 50 per cent of total transactions (being generally nearer the higher figure) C1 pp 59-61A, and that at Heckmondwike and Mirfield "no sales" were consistently around 28 or 29 per cent, but never reached 30 per cent, C1 pp 62-79. We also find that Mr Aleagha and Mr Mirzizad relied on employees to operate the Ossett shop, they themselves working at the other two shops.
- In evidence the appellants explained what happened to their takings and Z readings at the end of each evening. They said, and we accept, that the net takings from each till, net that is after the deduction of the wages of casual staff and moneys required, e.g. to buy locally food items needed in the shop, and the Z reading for the day were put into a paper bag. On the outside of the bag were written the gross takings and all deductions. Periodically details of the takings were transferred to a notebook in which the entries were made in Persian, the appellants' native language. Then, at the end of each VAT quarter, the entries in the notebook were transferred to three diaries - one for each shop - the diary entries being made in English. The Z readings were then destroyed. The diaries and the appellants' purchase invoices were given to their bookkeeper, Mr Patel, at the end of each quarter. Over a period of 2 or 3 days, he sorted the invoices by date and supplier, totalled them, and add-listed the takings. Mr Patel's work was then taken to the appellants' chartered accountants, Messrs George Barker & Co, where their records were completed and their VAT returns prepared by Mr Waide, the sole proprietor of the firm.
- The Appellants' VAT returns were signed by Mr Waide as authorised signatory. He also completed blank signed cheques provided by the appellants to cover their VAT liability.
- As Mr Lumb was aware that the appellants rendered VAT returns on a calendar quarterly basis, one of his concerns on 10 October 1998 was to establish the whereabouts of records that would form the basis of their 09/98 return. His enquiries of Mr Mirzizad indicated to him that the records were with the appellants' "accountant". (Strictly speaking, that reply was incorrect, and we deal with this aspect of the evidence in the next following paragraph). Consequently, early on the morning of Monday, 12 October 1998, Mr Lumb telephoned Mr Waide to ensure that he did have the records. Mr Waide said he had them up to 30 June 1998. Mr Lumb thereupon went to Mr Aleagha's home, where the two met. Mr Aleagha confirmed that he had that morning delivered the records to Mr Waide, having collected them the previous day from Mr Patel. Mr Lumb then went to Mr Waide's office and collected them. He had them in his possession by mid morning.
- As we have said, Mr Mirzizad told Mr Lumb that the books were with the accountant. He said they had been taken there 3 or 4 days earlier. But what he meant by that reply was that the books were with Mr Patel for him to carry out preliminary work before they were taken to Mr Waide. We are satisfied that Mr Mirzizad either made a simple mistake in replying to Mr Lumb or, perhaps due to language problems, misunderstood his question and gave an incorrect answer. But whichever was the case, in our judgment, there was nothing sinister in the incorrect reply. We are also satisfied that the work carried out by Mr Patel was on records completed by the appellants before 10 October 1998.
- However, Mr Lumb assumed that Mr Mirzizad had deliberately misled him, and that his purpose in doing so was to enable incorrect records to be prepared, or be completed, on which the appellants' 09/98 VAT return was to be, and was in fact, based. We find that Mr Lumb was not justified in making that assumption.
- The appellants' VAT returns for the periods 09/97 to 09/98 contained the following declarations (C1 pp 83A, 83C and 85):
| Period |
Sales at Mirfield and Heckmondwike |
Sales at Ossett |
Total Sales |
| 09/97 |
63440 |
- |
63440 |
| 12/97 |
64850 |
6361 |
71211 |
| 03/98 |
57422 |
8372 |
65794 |
| 06/98 |
59659 |
7085 |
66744 |
| 09/98 |
71710 |
16131 |
87841 |
- We find that, although there was an increase of 127 per cent in the declared takings at Ossett in period 09/98 when compared with period 06/98, the increase at Mirfield and Heckmondwike for the same period was of the order of 20 per cent. We also observe that, despite the declared turnover at Ossett in period 09/98 being more than double that in period 06/98, it was less than one quarter of the aggregate of that at Mirfield and Heckmondwike in the later period, i.e. even in the later period the Ossett turnover was less than half the turnover of either Mirfield or Heckmondwike, assuming (as we believe to have been the case) that the takings of those two shops were roughly equal.
- Mr Lumb, having concluded that the appellants had underdeclared their liability to tax, then proceeded to calculate the tax subsequently assessed on them.
- The method of calculation he adopted for the Mirfield and Heckmondwike shops involved a comparison between an estimate of the value of sales made between 2 and 8 March 1998 inclusive and the sales recorded for those dates. As the method required the estimation of several factors, to indicate just how many we propose to mark each step at which he made an estimate or assumption with an asterisk (*), coupled with its number.
- It will be recalled that Customs officers attempted to note the first and last purchases at Mirfield and Heckmondwike on each of the days in question in March. But that did not reveal the number of sales or their value. To obtain number and value Mr Lumb took the following steps.
- Firstly, in recognition of the fact that the appellants recorded a large number of "no sale" transactions on their till rolls, he had to estimate (*1) the total number of sales at Mirfield between Monday, 5 October 1998, and Saturday, 10 October 1998, starting by noting the "no sale" transactions. Mr Lumb's estimate is to be found at C1 pp 62-65.
- In the absence of a till roll for Sunday 4 October, he assumed (*2) that the sales on that day would have been an average of the figures for the remaining 6 days of that week. He estimated (*3) the number of "no sale" transactions at Heckmondwike on the basis of the till roll for Saturday, 10 October, and projected the figure produced using the Mirfield figures for the other six days, including the already estimated Sunday figure (C1 p. 81), assuming (*4) daily variations that mirrored exactly those at Heckmondwike.
- We find that in his calculations Mr Lumb did not arrive at an estimate of the "actual transactions", as he claimed (see C1 p. 62 et seq), for he ignored all non-transactional entries recorded as "no sales" (*5).
- Secondly, to estimate the average value of each transaction at Mirfield, Mr Lumb used the till rolls for Monday to Saturday (C1 pp 62-65), and calculated the figure (*2 repeated) for the Sunday as an average of the other six days of the week (C1 p, 73). He estimated (*6) the transaction values for Heckmondwike for Sunday to Friday using a combination of the Heckmondwike Saturday till roll and the Mirfield figures for Monday to Sunday (including the projected Sunday figure) to arrive, by use of percentages, at daily average transaction values (C1 p.80).
- Thirdly, Mr Lumb assumed (*7) that the daily October figures could be applied to the numbers of transactions counted in March to arrive at projected totals for the March takings (see C1 pp 66-72 for Mirfield and C1 pp 74-79 for Heckmondwike). He further assumed (*8) that the differences between the amounts declared for those days and the amounts projected represented under-recordings of sales.
- Mr Lumb thus concluded (see C1 p. 83 – Summary of sales projection from test purchase information) that, in relation to Mirfield and Heckmondwike, the true sales figure for the period between 2 March and 8 March 1998 inclusive was £6,325.78, as opposed to a declared figure of £4,708.10 – a suppression rate of 25.5 per cent. And, to quote from his schedule at (C1 p. 83A), "As constant trading pattern between declared sales and purchases, suppression rate applied to retrospective periods "i.e. from the appellants' commencement of trading in 1987 (*9).
- The fundamental underlying assumption Mr Lumb made (*10) was that the appellants systematically suppressed sales at a steady rate, so that evidence relating to the Mirfield or Heckmondwike shop could be applied to the other (*11), and that one period could be compared with another (*12). (We note that he did not think that to be the case at Ossett, although he gave no reason for not making the same assumption as to its operation.)
- As we have said, two of the witnesses for the Commissioners were the Tranmer brothers. Their firm supplied frozen burgers to the appellants. (The firm has now closed down). They were the only suppliers whose records were compared by Customs with those of the appellants, and it is common ground that the following Tranmer invoices were omitted from the appellants' records:
| Period |
Date |
Invoice No |
Amount |
| 12/93 |
23.10.93 |
6170 |
87.00 |
| 3/94 |
4.1.74 |
7681 |
87.00 |
| 6/94 |
1.4.94 |
9429 |
156.60 |
| 9/94 |
2.7.95 |
11272 |
87.00 |
| |
9.10.95 |
18765 |
52.20 |
| |
4.11.95 |
19193 |
121.80 |
| |
20.11.95 |
19413 |
52.20 |
| |
26.11.95 |
19504 |
87.00 |
| |
16.12.95 |
19809 |
121.80 |
| 12/95 |
30.12.95 |
19985 |
208.80 |
| |
5.2.96 |
20449 |
52.20 |
| |
10.2.96 |
20539 |
87.00 |
| |
19.2.96 |
20651 |
52.20 |
| |
16.3.96 |
21018 |
87.00 |
| |
18.3.96 |
21037 |
52.20 |
| 3/96 |
30.3.96 |
21178 |
69.60 |
| 6/96 |
29.5.96 |
21839 |
139.20 |
| 9/96 |
20.7.96 |
22465 |
174.00 |
| 12/96 |
21.12.96 |
24263 |
226.20 |
| |
15.3.97 |
25190 |
87.00 |
| |
12.4.97 |
25492 |
104.40 |
| |
5.4.97 |
25397 |
87.00 |
| |
3.5.97 |
25730 |
87.00 |
| |
24.5.97 |
25963 |
87.00 |
| 6/97 |
21.6.97 |
26279 |
87.00 |
| |
5.7.97 |
26445 |
121.80 |
| |
26.7.97 |
26671 |
87.00 |
| 9/97 |
23.8.97 |
26988 |
121.80 |
| |
16.10.97 |
27552 |
87.00 |
| |
24.10.97 |
27652 |
139.20 |
| |
22.11.97 |
27945 |
139.20 |
| |
6.12.97 |
28103 |
156.60 |
| |
17.12.97 |
28195 |
104.40 |
| 12/97 |
27.12.97 |
28304 |
104.40 |
| |
31.1.98 |
28641 |
104.40 |
| 3/98 |
21.2.98 |
28832 |
104.40 |
| |
15.4.98 |
29322 |
104.40 |
| 6/98 |
23.5.98 |
29698 |
104.40 |
- We observe that there were long periods when there were no omissions, e.g. from 9/94 to 12/95, and particularly note that the last Tranmer invoice omitted is dated 23 May 1998 i.e. over 4 months before the appellants became aware of Customs' interest in them. We regard that as significant.
- (Customs originally maintained that a further 7 invoices were missing, but they were later found either to have been recorded or to have been the subject of errors on the part of someone other than the appellants, e.g. misnumbering by Tranmer).
- Tranmers held a key to the appellants' Mirfield shop to which they made deliveries when it was closed. Their driver left the invoices for his deliveries with the burgers. The appellants claimed that that arrangement produced occasional mistakes. On one occasion an invoice for another of Tranmers' customers had been left with them and on other occasions their own staff had failed to give them invoices, and on yet others they claimed not to have received invoices. Those delivery arrangements were unique so far as the appellants were concerned, and we are satisfied that the appellants' claim that there were some mistakes is justified.
- It is common ground that the missing Tranmer invoices represented approximately 0.6 per cent by value of the appellants' purchase invoices for the whole of the various accounting periods they covered. We were not supplied with the Tranmer records for the whole of the period covered by the missing invoices. But for those periods which were covered we find that the appellants paid the missing invoices.
- Mr Lumb scheduled the appellants' purchases from Tranmer in periods 12/96 to 9/98 (including those for which the invoices were missing) at C1. p88 as follows:
| "Period |
Amount |
| 12/96 |
1757.30 |
| 3/97 |
1444.20 |
| 6/97 |
1809.60 |
| 9/97 |
1705.20 |
| 12/97 |
2314.20 |
| 3/98 |
1501.20 |
| 6/98 |
1687.80 |
| 9/98 |
1774.80" |
- From that analysis, he concluded (again C1 p88):
"1. No apparent increase of burger purchases in period 9/98 to support traders' claim that both purchases and sales increased in period 9/98 as a result of a sales promotion (menu drop).
2. Unrecorded burger purchases appear to have ceased after period 6/98 which supports our consideration that trader included more purchases and sales in period 9/98 records, provided after visit of 10.10.98."
- We do not accept his conclusions as fact.
- In evidence, Mr Lumb admitted, and we find, that Customs had made no attempt to check purchases from other suppliers. Thus the only evidence of missing purchase invoices adduced by the Commissioners related to purchases from Tranmer.
- Mr Lumb also claimed in evidence that, because the level of burger purchases by the appellants had not increased, but had remained constant to period 9/98, it was reasonable to assume that the appellants' purchases and sales had not increased by the amounts they claimed, i.e. as indicated by their VAT returns.
- However, in an analysis of the appellants' purchases carried out by his colleague, Mr Lewis, to be found at C1 pp. 90-100, the following information emerges:
| Supplier |
3/98 |
6/98 |
9/98 |
| Howarth |
1134 |
1137 |
1520 |
| T&L |
1286 |
1620 |
2652 |
| ICS |
5152 |
6787 |
7566 |
| Valerios |
7194 |
7301 |
7530 |
| Makro |
1213 |
1265 |
1573 |
| Ice Pak |
764 |
694 |
889 |
| Tranmer |
1322* |
1479** |
1775 |
| Woods |
390 |
450 |
480 |
| Cash |
558 |
662 |
1080 |
| Misc & Supermarket |
806 |
824 |
1043 |
| Sainsbury |
-
19819 |
-
22219 |
93
26201 |
*add unrecorded purchases of £208.80
** add unrecorded purchases of £104.40
- In reliance on Mr Lewis's analysis and of necessity the rejection of Mr Lumb's claim that there was no discernible increase in purchases, we find that the appellants' purchases increased by some 30 per cent from period 3/98 to 9/98. That the increase in burger purchases was much smaller – some 5.15 per cent – we consider irrelevant. It may well be accounted for, at least in part, by sales of kebabs having replaced those of burgers (see para 49 infra).
- We also observe that, with the exception of the "Cash" and "Misc & Supermarket" items (which formed but a very small part of the whole) , the purchases were made from identified traders who almost certainly kept full records.
- Based on the tax assessed in relation to Mirfield and Heckmondwike, the appellants' under-declared sales, expressed as a percentage of declared sales for the individual days of the week 2 March 1998 to 8 March 1998, ranged from 11 per cent at Heckmondwike on 8 March to 103 per cent at Heckmondwike on 3 March 1998. Further, they also gave rise to an over-declaration on 2 March at Heckmondwike.
- In relation to Ossett, Mr Lumb assumed that the appellants' declared sales in period 09/98 could be compared with those declared in period 03/98. His reasons for so assuming and his calculation of the percentage rate of suppression are to be found at Cl p. 83C as follows:
" — Ossett sales in period 9/98 (records produced for this period after visit of 10/10/98) = £16,131.10
- Consider more sales have been declared due to our known interest and not increase in custom
- Therefore as no test purchase information held for Ossett arrears assessed based on increase in sales declarations
- Previous highest declared sales for Ossett is in period 3/98 (£8372.50)
Therefore suppression rate [Diagram or picture not reproduced in HTML version - see original .rtf file to view diagram or picture] x 100 = 48%"
We shall deal with his reasons in our conclusion.
- The appellants had menus printed for the Mirfield shop in May 1998. They also had menus printed for Ossett in October, November and December 1997. They were invoiced by North Account Book Manufacturing Co for 5000 menus for Ossett on 20 October 1997 (P1.p.9), and for a further 9250 on 7 November 1997 (P1. p.11), and for yet another 10,000 on 12 December 1997 (P1 p.13)
- On 15 May 1998 the appellants were invoiced for 20,000 menus by Print Express (C1 p.16). Mr Gill maintained they were for Ossett, whilst the appellants both maintained they were for Mirfield and Heckmondwike. And on 12 June 1998 they were invoiced for 25,000 menus by Print Express (C1 p. 15). Mr Gill believed them to relate to Heckmondwike and Mirfield, whilst the appellants maintained they were for Ossett.
- But whichever party be correct in relation to supplies of the menus referred to in the last preceding paragraph, we find the minimum number of menus printed for Ossett totalled 44250. We also find that at least some of the Ossett menus included items not on the menu of the other two shops, kebabs being the most important of them.
- The appellants adduced evidence showing that a very large number of the Ossett menus were distributed by hand and that the Ossett area in which they made deliveries initially was extended in the hope of increasing sales. Both claimed that it was necessary to keep only a small number of menus in their shops. We accept their evidence as fact.
- They also claimed that at the outset of trading at Ossett their terms of delivery were less favourable than those of competing fast food outlets, so that they changed to free delivery. As no evidence of competitors' practices was adduced, we are unable to make a finding of fact as to their competitors' practices, but we should perhaps say that, viewed against all the information before us, the appellants' claim would appear to have foundation.
- In September 1998 the appellants installed a new telephone system at each of their shops. It included a facility to disclose the number of a person making an incoming call, and a memory for frequently used numbers. The appellants claimed, and we accept, that previously frequently used numbers were recorded on pieces of paper kept in till drawers, so that it was necessary to use the "no sale" key to obtain them, thus increasing the need for use of that key.
- In interview on 13 November 1998 (C1 pp.104-169), Mr Aleagha responded to certain questions put to him and Mr Mirzizad by Mr Gill in the presence of Mr Lumb about their bookkeeping and record system. That system, which we earlier described (see para 15 supra), was not one to be recommended for it failed to ensure that the necessary records were kept for the required period (see para 6.3 Schedule 11 to the Value Added Tax Act 1994). We find from the exchanges recorded that Customs were aware of the system adopted and used by the appellants, and that they either expressed satisfaction with it or accepted it notwithstanding their dissatisfaction. (Although it is not easy to discover precisely what Mr Aleagha was at times saying in interview, he claimed that a Mr Richardson, a Customs officer who made a control visit to the appellants some time previously, was told that they kept Z readings for only 6 weeks and then destroyed them. He also claimed that Mr Richardson subsequently informed the appellants that Customs were content to allow them to continue that practice. We particularly record that the appellants' evidence in interview about Mr Richardson's visits, as confirmed at the hearing, was not challenged by the Commissioners. In the absence of any challenge, we accept the claims as fact).
- And although by the date of the interview the Commissioners were aware that the appellants maintained that their books and records were with Mr Patel on 10 October 1998, Mr Gill (and for that matter Mr Lumb) chose not to raise any questions of them in interview as to Mr Patel's part in events. The appellants answered various questions put to them in a way that might have indicated a reluctance either to identify Mr Patel or to indicate the nature of his work for them. Yet no effort was made by the interviewing officers to challenge their claims.
- For completeness we add that, in interview, Mr Aleagha and Mr Mirzizad said that they were aware that they must declare all sales and purchases on their VAT returns, and that it was an offence knowingly to submit incorrect VAT declarations.
The Law
- The law in point in relation to the tax assessments is to be found in s. 73(1) and s. 77 of the Value Added Tax Act 1994. The relevant parts of the former subsection read as follows:
". . . [W]here it appears to the Commissioners that .. . returns [required under this Act] 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."
- And those parts of s. 77 which we must consider are the following
"(1) Subject to the following provisions of this section, an assessment under section 73 . . . shall not be made —
(a) more than 3 years after the end of the prescribed accounting period . . . concerned
(4) Subject to subsection (5) below, if VAT has been lost —
(a) as a result of conduct falling within section 60(1) . . ., or
(b) . . .
an assessment may be made as if, in section (1) above, each reference to 3 years were a reference to 20 years
(Subsection (5) is irrelevant in the present context.)
- As to the penalty assessment we must deal with sections 60 and 70 of the 1994 Act to the following extent:
"60(1) In any case where —
(a) for the purpose of evading VAT, a person does any act or omits to take any action, and
(b) his conduct involves dishonesty (whether or not it is such as to give rise to criminal liability),
he shall be liable . . . to a penalty equal to the amount of VAT evaded or, as the case may be, sough to be evaded, by his conduct.
(7) On an appeal against an assessment to a penalty under this section, the burden of proof as to the matters specified in subsection (1)(a) and (b) above shall be upon the Commissioners
70(1) Where a person is liable to a penalty under section 60 . . ., the Commissioners or, on appeal, a tribunal may reduce the penalty to such amount (including nil) as they think proper
(2) In the case of a penalty reduced by the Commissioners under subsection (1) above, a tribunal, on appeal relating to the penalty, may cancel the whole or any party of the reduction made to the Commissioners ."
Submissions and Conclusion
- Dealing first with the tax assessments, we proceed immediately to the question of quantum, there being no issue on best judgment, it having been abandoned as a ground of appeal by the appellants.
- In relation to quantum, we must throughout bear in mind two points on the assessments made by Mr Barlow, counsel for the appellants. First, he observed that the assessments were in the maximum sum the facts known to the Commissioners could justify. Secondly, he commented that, where it is alleged that appellants have underdeclared their liability to tax deliberately and therefore dishonestly, a best judgment assessment does not prove that there has been an underdeclaration; it merely asserts that an amount is due and leaves it to the appellants to disprove the assertion.
- We must also add that each assessment before us for an accounting period prior to 6/96 is valid only if the Commissioners can establish that the appellants dishonestly evaded tax in that period (see s. 77 Value Added Tax Act 1994).
- In relation to the tax assessments, Mr Poole, counsel for the Commissioners, advanced four core submissions. First, he contended that the extent of omission of Tranmer invoices from the appellants' records was significant, and far more likely to have been deliberate than to have been attributable to accident, oversight or honest error. Further, whilst accepting that the missing invoices represented 0.6 per cent by value of all purchases, he observed that for 1997 they represented 22 per cent by value of those from Tranmer and, for period 12/95, 34.58 per cent, again from Tranmer. He added that the records of other suppliers to the appellants had not been examined only because the Commissioners believed that the co-operation of those suppliers would not have been forthcoming.
- It is implicit in that addition that we are invited to assume that the appellants suppressed other purchases. We decline that invitation. There is nothing before us to warrant our finding that purchase invoices other than those of Tranmer were omitted by the appellants. We find that the only purchase invoices omitted by the appellants were the Tranmer ones listed at para 32 supra. Of Mr Poole's submission that the much higher rates of omission of the Tranmer purchase invoices in 1997 and period 12/95 were significant and indicative of deliberate behaviour, as opposed to accident, we need merely say that we accept that in periods 12/95, 3/96, 6/97 and 12/97 omissions approached an unacceptably high number. But those periods were but 4 of 19 in the total period of omissions, and in 4 of those periods there were no omissions whatsoever, and in 7 others there was but one. In each of the other 3 periods either 2 or 3 invoices were omitted. That, in our judgment, is to put the matter into perspective. We are unable to discern any pattern whatsoever in the omissions, and conclude that there was none. Looking at the complete picture, as we believe we should, we are simply unable to accept that the omission of 0.6 per cent by value of all purchase invoices could, in isolation, possibly justify a finding that the appellants suppressed sales at a rate of 25.5 per cent at Mirfield and Heckmondwike. We would expect to find considerable supporting or additional evidence to justify such a rate.
- Secondly, Mr Poole claimed that Mr Lumb's analysis of the appellants' sales transactions at Mirfield in March indicated that they had declared 430 sales, whereas they had effected about 600; and since there was no evidence to show that the March period was other than representative, he submitted that there was a significant unexplained underdeclaration of sales at Mirfield, and for that matter at Heckmondwike. He explained how the figures of 430 and 600 had been arrived at in this way. In the week commencing 2 March 1998 the appellants declared sales at Mirfield totalled £3463.50, and the Commissioners' 'first and last' purchases indicated that they recorded 857 transactions of which about 600 (70 per cent) represented true sales. Using Mr Lumb's analysis of the appellants' records at C1 pp 66-73 , their average transaction price was £5.73. Applying that price to the sum declared indicated 430 sales.
- Mr Poole did however acknowledge that Mr Lumb had erred in dividing total sales only into "actual transactions" and "no sales transactions" so that he had failed to reduce the figure for the former category to take account of returns, voids, and various other genuine entries. Consequently Mr Poole accepted that " the labelling in Mr Lumb's schedules might be open to criticism". Nonetheless, he submitted that it made no difference to the outcome of the assessments: they indicated a significant unexplained underdeclaration of sales.
- Assuming the figures advanced by Mr Poole to be justifiable, we have no means of knowing, or even calculating in the roughest terms, the true extent of Mr Lumb's error: nor, in the absence of evidence on the point, have the appellants. Coupled with the error, we must take account of the considerable number of assumptions in his analysis which, particularly in combination, have the ability very considerably to distort the takings figures he calculated. Whilst the burden of proof in relation to the tax assessments rests on the appellants, we are dealing with a case where the majority of those assessments have validity only if the Commissioners can prove that the appellants dishonestly evaded tax (see para 61 supra). We would therefore expect, indeed require, the whole case against them to be particularly strong. To take account of that requirement we proceed on the basis that reconciliation of the figures of 600 actual transactions calculated and 430 sales apparently declared is possible unless our further deliberations indicate that it is not. We are not prepared to consider the differences in the two figures to be the supporting or additional evidence we indicated we should seek in para 63 supra.
- The third core submission made by Mr Poole was that, without concession, the appellants had maintained that they had declared all their sales, yet
1) as the Z readings were disposed of before inspection by their bookkeeper and accountant, there was no means of verifying their sales records; and
2) investigations by Customs had indicated their sales records to be unreliable and inaccurate.
- Earlier (see para 53) we found that the Commissioners by Mr Richardson either expressed their satisfaction with the appellants' recording method or, notwithstanding that they were dissatisfied with it, nevertheless indicated that the appellants might continue to use it. And, as we also found, nothing in the evidence as to events either in March 1998 or on 10 October 1998 indicated that the appellants were making untrue sales declarations. No other evidence was adduced that might have satisfied us to the contrary. We find that the records produced by the appellants to Customs on 12 October 1998 were reliable and accurate.
- It will be recalled that Mr Lumb concluded that the appellants underdeclared takings throughout their period of trading in relation to both Mirfield and Heckmondwike "as constant trading pattern between declared sales and purchases". Consequently, he applied the suppression rate of 25.5 per cent to "retrospective periods" i.e. back to 1987 (C1 p.83A). In evidence he added that he had analysed the ratio of sales: purchases quarterly, and did not consider the trading pattern to be an unreliable basis of assessment. We understood him to mean that he considered the ratio of sales:purchases to have been relatively constant throughout the period of trading and sales to have gradually increased, so that it was not unreasonable to assume that the appellants had been systematically suppressing takings at a steady rate throughout the period.
- Included in Mr Lumb's working papers is a "Schedule of outputs, inputs and VAT declared" (C1 p. 89). From it we select at random (as being more or less in the middle of the assessment period) the outputs and inputs figures for the four accounting periods ending in 1992. They read:
VAT quarter Outputs Inputs
March 1992 18622 9218
June 1992 26654 13602
September 1992 27837 22070
December 1992 35476 24880
- We observe that, in round figures, inputs were ½ the outputs in the two earliest periods, were ¾ in the third period, and 5/7 in the fourth. That hardly reflects a constant trading pattern. Further, the turnover in the December quarter was almost double that in the March quarter – a matter that might reasonably have given the Commissioners cause for concern. Despite the most careful consideration of the whole of Mr Lumb's schedule at C1 p 89, we are unable to accept either that there was a "constant trading pattern between declared sales and purchases" in the assessment period, or that declared sales increased in such a way as to indicate a systematic pattern of suppression
- We must also deal with the claim by the Commissioners that "dramatic" increases in both sales and purchases in period 9/98 indicated suppression of takings. We earlier found that declared sales at Mirfield and Heckmondwike in that period increased by 20 per cent over those in the immediately preceding period (see paras 20 and 21 supra), and that total purchases in the periods 3/98, 6/98 and 9/98 increased by 30 per cent overall (see paras 42 and 43). All we need say is that we are satisfied on the evidence that the substantial increases in both declared sales and purchases by the appellants in period 9/98 were genuine.
- Nothing in paras 67 to 72 of our decision provides the supporting or additional evidence we earlier said we should seek to support a suppression rate of 25.5 per cent. We are unable to accept that the appellants suppressed takings at Mirfield and Heckmondwike at that rate. Indeed, we go further and say that on the evidence the appellants have satisfied us that there were no underdeclarations of takings at the two shops in the assessment period. We reduce the assessments in respect of the two shops to nil.
- Mr Poole's fourth core submission was to the effect that suggestions by the appellants that the large increase in sales at Ossett in period 09/98 was due to factors such as a menu drop were not credible.
- That the Ossett takings rose by 127 per cent between periods 06/98 and 09/98 (from £7085 to £16,131) we accept and, whilst that increase may be described as "dramatic", we are satisfied that it is accounted for by Ossett being a newly opened shop and by the efforts made by the appellants inter alia both to publicise its existence by menu drops and to attract customers by offering a delivery service. It greatly concerns us that, whilst the Commissioners allege that there was steady and systematic suppression of takings at Mirfield and Heckmondwike at a rate of 25.5 per cent, in respect of the newly opened shop at Ossett they allege suppression at 48 per cent based on nothing more than a declared increase in takings of that amount over a six month period. We would expect a "dramatic" increase in turnover from a newly opened branch of an existing, successful business, and in the absence of any evidence to the contrary we find that the increase was genuine.
- Mr Lumb dismissed the appellants' drop of some 45000 menus at Ossett as having accounted, at least in part, for an increase in sales. He did so originally on the basis that a menu drop in November 1997 took place at Mirfield and produced no increase in sales there. As that menu drop was at Ossett – a fact he accepted before us – he made a serious error, which he appeared to be unwilling to correct.
- We dismiss a claim by the Commissioners that had there been a substantial increase in turnover at Ossett it would necessarily have been reflected in the number of staff employed there. As Mr Barlow pointed out, before an increase in trade requires additional staff, it is logically necessary to show that the existing staff are fully occupied. If the staff are only working at half capacity, turnover can double without any increase in wages. No evidence was adduced by the Commissioners to indicate the necessity for additional staff.
- Beyond the fact that purchases for Ossett were included in the appellants' total purchases, so that that shop's operation might have been tainted by association with the other two, there is not a shred of evidence to indicate that the appellants suppressed takings at Ossett. Mr Lumb's statement at C1 p. 83C (see para 46) that "more sales have been declared due to our known interest and not increase in custom" is totally unsupported by evidence, and we reject it as justification for the Ossett assessments. If ever assessments justified the description capricious, they were those made by the Commissioners in respect of Ossett. We are satisfied on the evidence that the Ossett assessments should be reduced to nil, and we so reduce them.
- The tax assessments under appeal having all been reduced to nil, the penalty assessment falls with them. We formally allow the appeal against it.
- We allow the appeal in its entirety and direct the Commissioners to pay the appellants' costs of the appeal, to be assessed if not agreed.
- Had it been necessary for us to consider the question of the dishonest evasion of tax, we should have held that in none of the accounting periods covered by the tax assessments did the appellants behave dishonestly. Consequently, we should also have held that all the assessments for periods prior to that of 6/96 were invalid. And under no circumstances should we have held that the penalty assessment relating to Ossett was merited.
DAVID DEMACK
CHAIRMAN
RELEASE DATE:
MAN/99/361
