[Home] [Databases] [World Law] [Multidatabase Search] [Help] [Feedback]
	United Kingdom VAT & Duties Tribunals Decisions
You are here: BAILII >> Databases >> United Kingdom VAT & Duties Tribunals Decisions >> Hung (t/a Hong Kong) v Customs and Excise [2004] UKVAT V18664 (21 June 2004) URL: http://www.bailii.org/uk/cases/UKVAT/2004/V18664.html Cite as: [2004] UKVAT V18664

[New search] [Printable RTF version] [Help]

18664

Value Added Tax - Section 73(1) VATA 1994 – assessment made to best judgment - method of calculating under-declared VAT – whether method used by Commissioners was so unreasonable that resulting assessment was not made to the best of their judgment – no – whether additional factors not made known by Appellant to Commissioners should be taken into account so as to reduce the assessment – yes – appeal dismissed – directions to reduce the assessment

LONDON TRIBUNAL CENTRE

CHI HUNG WAN trading as HONG KONG Appellant

and –

THE COMMISSIONERS OF CUSTOMS AND EXCISE Respondents

Tribunal: MR A E SADLER (Chairman)

MRS R A WATTS-DAVIES, MHCIMA FCIPD

Sitting in public in London on 15 and 16 April 2004

Mr Alex Ng of Counsel and Mr A Lau, accountant, instructed by Messrs Andrew Cross Lau, chartered accountants, for the Appellant

Mr Robert Keller of Counsel, instructed by the Solicitor for the Customs & Excise, for the Respondents

© CROWN COPYRIGHT 2004


 


DECISION

The Appeal and the Decision

1. This is an appeal by Mr Chi Hung Wan who at the material times carried on a Chinese takeaway food business, trading as Hong Kong ("the Appellant"). By an assessment dated 26 June 2002 the Appellant was assessed to VAT in the sum of £15,595.00 together with interest on that amount of £1,431.78, and the Appellant appeals against that assessment.
2. The Commissioners of Customs & Excise ("the Commissioners") raised their assessment under the provisions of Section 73(1) of the Value Added Tax Act 1994 ("VATA") on the grounds that the Appellant had under-declared the amount of VAT payable by him for the period beginning with the date on which he registered for VAT (8 June 2000) and ending on 31 October 2001. The Commissioners used their powers to assess the amount of VAT due from the Appellant to the best of their judgment.
3. The Appellant argues that the assessment should be discharged on the grounds that it was not made to best judgment since it is unjust and incorrect in amount, being based on estimates of sales made by the Appellant which are excessive, having been derived from observations of transactions of the Appellant's business where the method of observation was inappropriate.
4. It is our decision that the assessment was made to the best of their judgment by the Commissioners. It is also our decision that certain relevant factors which were put to us in evidence were not taken into account by the Commissioners in arriving at their calculation of the under-declared VAT, and that in consequence the amount of VAT assessed is excessive. We therefore direct that the amount of VAT should be recalculated in the manner we specify below, and that the assessment should be reduced to the amount resulting from such recalculation (including a corresponding reduction in the amount of interest).

The evidence and the basis of calculation used for the assessment

5. We heard evidence from Mr S.F. James, the officer of the Commissioners who was responsible for the visits to the Appellant's premises and the observations of sales which were made there; from Mr C.M. Scott, also an officer of the Commissioners who took part in the visits and who is the officer responsible for the calculations which resulted in the assessment; and from the Appellant. We also had submitted to us witness statements made by the various officers of the Commissioners who, under the supervision of Mr James, had assisted him in the observations of sales made at the Appellant's premises – these officers did not appear at the hearing, and the Appellant did not seek to challenge their evidence as set out in their witness statements.
6. The Commissioners produced a bundle of documents which included the detailed results of the observations made by Mr James and his team on their various visits to the Appellant's premises; the initial figures (based on these results) put to the Appellant as the likely amount of under-declared VAT; the correspondence between the Commissioners and the Appellant's accountants; and the letter from Mr Scott setting out the basis on which the amounts actually assessed were calculated. At the hearing the Commissioners submitted the briefing and instructions note given to the officers carrying out the observations at the Appellant's premises on 8 February 2001; they also submitted a chart prepared by Mr Scott to illustrate diagrammatically the way in which he had calculated the VAT assessed as the amount under-declared by the Appellant; and the Appellant submitted a diagram map showing the location of his premises in relation to street and car park parking facilities.
7. We accept without reservation the evidence of the Commissioners' officers and also the evidence of the Appellant, and from that evidence, and from the documents before us, we find the following facts.
8. The Appellant carried on business supplying takeaway Chinese food from premises at 1 Kingsway, Hemsby, Great Yarmouth, Norfolk. He had previously been employed as a chef in Chinese restaurants or takeaways, and in total he has had ten years' experience as a chef. For approximately one and a half years the Appellant was employed as a chef by the former proprietor of the business, but in June 2000 he acquired the business, and was registered for VAT purposes as a sole proprietor with effect from 8 June 2000. Since the events and periods which are the subject of this appeal, the Appellant has ceased to carry on the business and has de-registered for VAT purposes. This was the first business venture carried on by the Appellant as proprietor.
9. The business is located in a town which is a seaside resort, and it is near a holiday camp. The business is therefore seasonal, and its busiest times are during the summer holiday periods (especially late July through to early September). In the winter months the business is dependent upon local customers. The business has an evening trade only: it is open from 5pm until midnight, seven days a week (although sometimes it will open late – at about 5.30pm and will close before midnight). In the summer period its busiest time of trading is from 6pm to 9pm, and in winter from 5pm to 7.30pm. In August, however, trading is busy throughout the whole of the opening period.
10. The kitchen of the premises has four gas rings, permitting two chefs to work with two woks each. There is a separate small gas cooker which is used only for sauces. There is a small fryer and a large fryer for deep fat frying. All food is freshly cooked to order. Some dishes can be cooked in 4 to 5 minutes, but some require up to 10 minutes of cooking.
11. From the commencement of the business in June 2000 until the Easter holiday in 2001 (mid-April) the Appellant was the only experienced chef working in the business. He was assisted throughout this period by a Mrs Chung (the landlady of the premises) who was not an experienced chef, but who did the deep frying, cooked certain rice dishes, prepared any cold dishes, and packed the cooked food for customers. The Appellant also employed, on a part-time basis, four or five young people aged between 14 and 21 three of whom are children of Mrs Chung, and the others their friends. The Appellant and Mrs Chung worked only in the kitchen. One or more of the young people worked at the counter, taking orders from customers (including telephone orders), handing over the cooked and packaged dishes, and taking payment (cash or cheques) for the orders.
12. At the height of the summer season, in his first year, the Appellant struggled with the volume of business. At peak periods, as the only experienced chef, he could not deal satisfactorily with all the orders, and there were complaints about the waiting time for orders, orders were cancelled and money was refunded for some pre-paid orders where the customers were dissatisfied. These factors had a negative impact on the business and on the reputation of the business.
13. To avoid these problems in his second year, the Appellant sought additional experienced help in the kitchen. His father and other members of his family, who are experienced chefs working in London, came to help him out, so that from Easter 2001, for weekdays during the main holiday periods, and at week-ends and Bank Holidays throughout the whole of the summer period, there were two experienced chefs at work in the kitchen (the Appellant plus one of his family), whilst Mrs Chung continued with the jobs she was able to do, as described above. No changes were made to the staffing arrangements at the counter: Mrs Chung's children and their friends continued to take the orders and payment. The business was more organised with the additional chef, the problems experienced in the first year were largely eliminated, and its reputation improved.
14. The only other change identified by the Appellant which might have had an impact on the levels of business was that at some time in the winter of 2000/01 another Chinese takeaway business in the town, which could be regarded as a competitor business, closed down. No evidence was given as to how or whether this did in fact affect the levels of the Appellant's business.
15. The system for orders and payment, as operated by the young people employed at the counter, was as follows: customers visited the premises and placed orders, which were written down by the staff on unnumbered meal tickets (sometimes referred to as "kitchen bills") and handed through to the kitchen. Usually, but not always, the meal tickets were put on a spike by the kitchen staff. Telephone orders were dealt with in the same manner. Payment (mostly cash, sometimes cheques) was made when the food was handed to the customer.
16. In the early months of the business (including the date of the August 2000 observation visit by officers of the Commissioners, as described below) the Appellant had no proper till – he had borrowed an old till, but it didn't work, and therefore during this period cash payments were simply placed in a drawer under the counter. There was therefore no record of payment against orders. The day's takings were written down each night by the Appellant, and these figures were then used in preparing weekly turnover sheets, which in turn were used in preparing the VAT returns – it is not clear whether in this initial period the Appellant had any system for checking the day's takings against the day's meal tickets.
17. By Christmas 2000 the Appellant had acquired a till which worked. The till was kept out of sight of the counter and customers (including the Commissioners' officers in their later visits) could hear an amount being rung up on a till, but could not see if that related to their purchase, nor, of course, could they see whether the amount rung up corresponded with the price paid for the food. Payment for certain small items priced at less than £1 (such as a sachet of curry sauce, or plastic cutlery) was not recorded on the till at the time of purchase – instead, the cash payment was put in a drawer under the counter, and at the end of the evening the Appellant made a block entry in the till for the amount of cash in the drawer. The routine of the Appellant at the end of each day's business was to count the money in the till, check the total cash against the total amount shown as entered on the till that day for payment for food, and make a check of the total meal tickets against the till roll entries. Sometimes there would be a small discrepancy in the amount of cash as against the total takings recorded by the till roll. The meal tickets would not necessarily tally with the till records, since telephone orders were not always collected. The total turnover for the day (the till roll figure) was entered on a weekly sheet, and the weekly sheets were sent by the Appellant to his accountants for them to use in completing the quarterly VAT return.
18. There was no close supervision of the young people who worked at the counter – the Appellant was engaged full time in the kitchen. Nor did the Appellant carry out any spot checks to see if they were applying the correct procedures for taking and recording payments. He knew them well and trusted them.
19. The records of the business (meal tickets and daily statements of takings) for the period June to September 2000 were lost at the time of some refurbishment of the premises in the winter of 2000/01. Accordingly there were no records of the business against which the observations made in the visit by the Commissioners' officers on 11 August 2000 could be compared.
20. In correspondence with the Commissioners, the Appellant's accountants stated: "[the Appellant] denied any suppressions of takings but at the same time did not rule out any errors that the counter staff may have made during busy hours of the transactions." At a later stage in the correspondence they stated: "[the Appellant] would accept the limit of error to a maximum of 5% plus or minus but never agrees with any of the figures as estimated by you."
21. For the week of 11 August 2001 (that is, the week of the anniversary of the first visit by Commissioners' officers to the Appellant's business), the Appellant's declared weekly takings were £8,503.
22. On 11 August 2000 (a Friday) Mr James organised a series of test purchases by a number of Commissioners' officers in a covert operation at the Appellant's premises. The number of officers required for the visit was such that a special van or minibus had to be used. Street parking restrictions were such that it could not be parked in a street which gave a view of the Appellant's premises. There is a public car park from part of which it would be possible to have a view of the premises, but a bar over the entrance prevents larger vehicles (including the van or minibus) from using the car park. Therefore the officers could not carry out observations of all customers entering the Appellant's premises in the course of that evening's trading: instead a sequence of different officers made a series of test purchases at intervals throughout the evening. The first such purchase was made at 16.55hrs, and the last, by Mr James himself, at 23.30hrs. Seven such purchases were made in total at about hourly intervals. We were shown a copy of the instructions to the officers for the February 2001 visit (detailed below), but not for the August 2000 visit, but we assume that the instructions for the August 2000 visit were in similar terms. The instructions we were shown included the following : "Most important – Make sure that you time your entrance to obtain at least one overheard, and if possible note any others that you hear while you are there." An "overheard" is a purchase made by another customer (not an officer) whilst the officer is on the premises, where the officer overhears (and notes in his records) all or some of the details of the transaction. Thus the visits of the officers were not made to a strict timetable regardless of other factors – officers worked to an outline timetable or rota (for the most part visits on the hour throughout the trading period), to stagger their visits, but would wait until they knew that there would be other customers present, in the hope that they could have a record not only of their own purchases, but also of a number of "overheards".
23. From the notes of the officers' visits, recording their own purchases and the overheard purchases which they witnessed, Mr James compiled an estimate of the evening's takings for 11 August 2000, using a number of assumptions. First, Mr James noted the number of "overheards" where the officer had noted down the price paid for the order (28), and calculated the average price of each of those "overheards" (£14.18). This average price was then taken as the assumed price of each "overheard" where the purchase had been witnessed by an officer, but where he had no note of the price for that transaction. Mr James then aggregated the value (price paid) of all the actual purchases made by the officers with the value of all the "overheards" (using the actual prices paid where noted, and the average assumed price in those cases where the actual price had not been noted). This aggregate amount was £559. Next, Mr James compared the aggregate time he and his officers spent in the premises that evening (the "time in" - 1 hr 24 mins) with the aggregate time from the first noted transaction at 16.55 to the last noted transaction at 23.40 when no officer was present (the "time out" - 5 hrs 21 mins). He assumed that the rate of business throughout the "time out" was (at least in terms of the value of food sold) the same as the rate of business throughout the "time in": this calculation gave him estimated takings for the whole evening of £2,695.42. He then made a further adjustment: he noted that, even on the basis of the noted transactions, there was not in fact an even spread of customers over the whole evening. He himself was on duty to make the final visit, and his observation was that custom tailed off. He therefore readjusted his calculation so as to disregard all "time in" and (more significantly) "time out" after 22.16 (the time of the last "overheard" noted by the officer making the penultimate visit). With this adjustment the calculation of estimated takings for the whole evening was £1,866.02.
24. Mr James and his team of officers carried out a similar exercise of test purchases and "overheard" transactions, conducted on similar lines, on 8 February 2001 (a Thursday). The same methods of observation and the same methods of calculation (and the same assumptions) were used as with the August exercise, and again the adjustment was made in the ratio of "time in" to "time out" so as to disregard the period after the penultimate visit by an officer. On this occasion the calculation of estimated takings for the whole evening was £757.47.
25. A further visit was made by Mr James and some colleagues on 7 September 2001, but on this occasion (at least on the evidence before us) there was no such calculation of estimated takings. At the end of that visit Mr James made himself known to a member of staff on the premises at the close of business (the Appellant was not present), and arrangements were subsequently made for a meeting with the Appellant so that his records and declarations of sales could be examined and compared with the observations and calculations made by Mr James and his colleagues.
26. For the reasons given above, the Appellant was not able to produce detailed information as to the takings of the business on 11 August 2000 – all he could provide was the total declared takings for that evening of £541.40. This compares with Mr James's calculation of estimated takings of £1,866.02. Further, the total value of purchases made that day by Mr James and his colleagues and the "overheard" transactions where the price was also noted (i.e. excluding the 8 "overheards" where the price was not heard) was £476.45, which is 88% of the declared takings – but officers had been present in the premises for only approximately 20% of the period of trading.
27. As for the visit on 8 February 2001, the Appellant was able to produce the day's till role detailing each entry, and from which he had taken the total takings for the day as subsequently declared in his VAT return. Comparing the till roll entries with the observations made by the Commissioners' officers, although all of the actual purchases made by the officers that evening were properly recorded on the till roll, 3 of the "overheards" were not recorded. The total takings recorded on the till roll were £286.05, compared with Mr James's calculation of estimated takings of £757.47. The total value of purchases made that evening by the officers and the "overheards" where the price was also noted (i.e. excluding the 5 "overheards" where the price was not heard) was £191.08, which is 66% of the declared takings – officers had been present on the premises for approximately 23% of the period of trading.
28. Mr James also visited the Appellant's premises during trading on 5 February 2001 (in preparation for the major observation exercise which took place on 8 February 2001). Mr James made a purchase and observed 4 other "overheard" transactions where he was able to note the price charged. When Mr James compared the till roll for that evening with his observations he noted that only 3 of the 5 transactions were shown on the till roll.
29. In relation to the visit on 7 September 2001, the meal tickets or kitchen bills (on which the orders were taken by the counter staff and handed through to the kitchen) were compared to the till roll. Some of the kitchen bills had been "spiked", and some not. Some of the "overheard" transactions did not appear on the till roll.
30. On this information the Commissioners concluded that the Appellant had under-declared his takings when making his VAT returns. The internal "Report on Visit to Trader" prepared by the Commissioners notes in detail the results of the visits and the observations made and concludes that the 11 August 2000 visit suggests a suppression rate of 70.9%, and the 8 February 2001 visit suggests a suppression rate of 62.9%. A suppression rate of 70.9% requires that declared takings are uplifted by 344% in order to reach the "actual" takings; a suppression rate of 62.9% requires declared takings to be uplifted by 264%.
31. That Report also makes reference to the Appellant's declared takings as against the declared takings of the immediately preceding business carried on at the premises, as follows: "The new trader's [i.e. the Appellant's] first return covered the obs day [i.e. 11 August 2000] – he had declared considerably more that [sic] the previous trader (with no obvious changes, other than increased competition in the area) but a new calculation still indicated a significant amount of suppression." Elsewhere in the Report it is stated that during this period the Appellant had declared about 50% more than was declared by the previous trader.
32. On 6 December 2001 Mr James wrote to the Appellant to inform him that the Commissioners considered that there had been an under-declaration of amounts shown on his VAT returns, to set out some preliminary calculations of the true levels of takings based on the observation visits, and to invite the Appellant to provide any information which might result in a more accurate statement of actual sales for the relevant period. No assessment was made at this stage, but the preliminary calculations given in the schedule to the letter were stated to show "the amount of tax due if I issue an assessment based on this figure". The schedule covers the period from the commencement of trading on 8 June 2000 until the end of July 2001 (4 quarterly VAT periods, included the initial extended period). Two sets of figures are given, the first based on an uplift rate of 262% (i.e. assuming a suppression rate of just under 62%), and the second based on an uplift rate of 264% (assumed suppression rate of just over 62%). Taking the second set of figures, declared takings (inclusive of VAT) for the period totalling £191,091 are increased (again, inclusive of VAT) to £504,484, and the tax payable on such assumed sales (£75,135) is compared with the tax paid on declared takings (£28,460), giving a tax under-declaration of £46,675.
33. There followed correspondence and meetings with the Appellant's accountant advisers. They challenged the figures calculated by Mr James on the grounds that they implied average weekly gross takings of £8,485, a level of sales out of all proportion to the actual business, and far in excess of average takings of small takeaway food businesses, in their experience. They challenged the basis of the method of observation, in particular the assumption that the rate of business was constant during opening times, and that the same rate should be used for the "time out" periods as for the "time in" periods in circumstances where officers had, in accordance with instructions, waited until there were known to be customers in the premises before entering to make their purchases and observe other transactions. They pointed out that in the visit on 8 February 2001 the till roll showed that all the actual purchases made by officers had been shown to have been properly recorded on the till roll and therefore included in the declared takings. At this point neither the Appellant nor his accountants offered any further information or figures which might lead to any alternative calculation of total sales, except that they suggested that, based on the average price per dish and average cooking times, the maximum output per chef would be sales of £420 (this basis of calculation was elaborated in correspondence after the assessment was made, as mentioned below).
34. The matter was reconsidered by Mr James together with his colleague, Mr Scott. They concluded that the information obtained from the various visits showed that takings had been under-declared because all sales had not been recorded, but that this was not a case of deliberate suppression of takings by the Appellant. They concluded that they had grounds for making an assessment to "best judgment" to recover the tax on under-declared takings. They also concluded that an assessment to "best judgment" should not be based on the calculations previously made by Mr James – although they did not expressly accept any of the criticisms made by the Appellant's accountants as to the method of observation and the assumptions used in calculating the figures originally put to the Appellant by Mr James, they clearly felt that they did not represent a fair and sustainable estimate of the likely under-declaration: as Mr Scott put it in his evidence, the observations and calculations made by Mr James opened their eyes to the fact of suppression, but did not provide empirical evidence of the likely amount of suppression. Their difficulty was that no further information was provided by the Appellant as a basis either for revising the original calculation or for adopting an alternative method of calculation.
35. Mr Scott therefore decided to use a different method, and produced a different set of calculations. Since the figures resulting from his calculations were the figures which appeared in the eventual assessment it is necessary to set out his method in some detail.

i. His benchmark was the level of sales declared by the Appellant following the date (7 September 2001) when the Appellant knew that he was under investigation – from these it was clear that the business could achieve sales of approximately £8,000 per week, at least during its busiest quarter to the end of October.

ii. He began by taking the declared average weekly sales figure for the VAT quarter 10/01 (approximately £6,000) and increased that by 20% to give assumed actual weekly sales for that quarter of £7,200 (giving an effective suppression of sales at the rate of 16.5%). The uplift of 20% was not based on Mr James's calculations (which had given a very much higher rate of uplift – 264%) – it was chosen, in the absence of any other evidence, as a credible increase which gave a result within the parameters of the level of trade which the business was capable of sustaining, by reference to the "benchmark" of declared sales of £8,000 per week.

iii. The same uplift of 20% was applied to the declared sales for the previous VAT quarter, 07/01, so that declared average weekly sales of £5,000 were re-stated at £6,000 for that quarter.

iv. Next, Mr Scott looked at the average weekly declared sales for the VAT quarter 01 /02, and assumed them to be correct (i.e. no suppression) as they fell after the date the Appellant knew of the investigation: they were £2,875; he compared them with the average weekly declared sales for 01 /01 of £1,982. He assumed (in the Appellant's favour) a 10% price increase between 2001 and 2002, but he also assumed that levels of business in the 01 /01 quarter were the same as in the 01 /02 quarter. After allowing for the assumed price increases, an uplift of 30% is required to bring the 01 /01 average weekly sales figure to that of 01 /02, and this is assumed in Mr Scott's calculations for assumed sales throughout the 01 /01 quarter.

v. For the quarter 04/01 Mr Scott did not have recourse to "correct" sales for the corresponding quarter of 04/02. Therefore he assumed that an uplift of 25% on declared sales was a fair assumption to arrive at actual sales – 25% being the mid-point between the 30% uplift used for the preceding 01 /01 quarter and the 20% uplift used for the following 07/01 quarter.

vi. This leaves the initial quarter 10/00 – in fact the 21 weeks from the commencement of business on 8 June 2000 to 31 October 2000. Here Mr Scott notionally divided it into an 8 week period (to the end of July 00) and a "normal" quarter to the end of October 00 – this permitted him to make direct comparisons with the 07/01 and the 10/01 quarters respectively. In summary he did the following: he assumed a 10% price increase between the corresponding 00 and 01 quarters, but otherwise assumed that the levels of business in the 00 quarters were the same as in the corresponding 01 quarters; he had calculated (see (iii) above) that the actual average weekly sales for 07/01 were £6,000, and he assumed that (with allowance for the notional price increase) average weekly sales for the 8 weeks to end of July 00 were also at that level; likewise, he had calculated (see (ii) above) that the actual average weekly sales for 10/01 were £7,200, and he assumed that (again, with allowance for the notional price increase) average weekly sales for the 13 weeks to end of October 00 were also at that level. Re-stating the sales figures for the 21 week period to end of October 00 on this basis gave an uplift of 190% on the declared sales for that period, which Mr Scott felt was comfortably within the scope of the 264% uplift derived from the calculations produced by Mr James.

36. The result of Mr Scott's calculations was an under-declaration of VAT by the Appellant of £9,171 for the 21 weeks in the "quarter" 10/00; £1,151 for the quarter 01/ 01; £1,243 for the quarter 04/01; £1,912 for the quarter 07/01 and £2,218 for the quarter 10/01. This totals £15,695.
37. On 22 May 2002 Mr Scott wrote to the Appellant's accountants setting out his basis of calculation and the resulting amounts of VAT which the Commissioners considered to be under-declared, and this was followed by the assessment dated 26 June 2002 for the amounts set out in Mr Scott's letter, as detailed in paragraph 36 above, plus interest of £1,431.78.
38. By this time the Appellant had ceased to trade. His accountants asked for the matter to be reviewed, and it was referred to the Cambridge & Norfolk Review and Appeal Team of the Commissioners. In the course of the correspondence which ensued, the Appellant's accountants produced a schedule referred to as a "Cooking Capacity Exercise". This used information from the observations made by Mr James and his colleagues on 11 August 2000 and 8 February 2001 to show that each dish ordered took on average nearly 6 minutes to prepare. However, the Appellant was prepared to assume that, at full capacity, an experienced chef could prepare a dish in 4 minutes (15 dishes an hour). If the average price per dish is £4, then an experienced chef can generate sales of £60 (15 x £4) per hour, or £420 for a 7 hour evening. If it is assumed that there are two experienced chefs, and an assistant preparing deep fried dishes and cold dishes (the role of Mrs Chung) who generates sales at the same level, then the maximum daily output of the business is £1,260 (3 x £420), and the maximum weekly output is sales of £8,820 (7 x £1,260).
39. The Commissioners did not consider that these figures as to maximum capacity threw into question the basis of their own calculations, and so the assessments were not withdrawn or amended, and the matter proceeded by appeal to this Tribunal.

The submissions

40. For the Appellant, Mr Ng and Mr Lau asked us to find that the assessment was so flawed that it could not be regarded as made to the best of their judgment by the Commissioners and should therefore be dismissed.
41. First, they challenged the methods used by Mr James and his team and the assumptions then applied by Mr James in making his calculations. In particular they challenged the assumption that the levels of custom throughout the evening should be regarded as constant, and they argued that the calculations were distorted because officers had arranged their visits to coincide with periods when other customers were present and it was then assumed that the "time out" periods were as consistently busy as the "time in" periods of observation. They pointed out that the calculations by Mr James made no allowance for seasonal variations. They also pointed out that, in the 8 February 2001 observation, all of the 9 test purchases made by the officers were properly recorded and included in the declared takings, and that with "overheard" transactions there was room for uncertainty or question. They argued that the initial figures put forward by Mr James as under-declared VAT were clearly massively overstated, and were used to frighten the Appellant into providing information, rather than as a genuine attempt to determine any unpaid VAT.
42. Secondly, they challenged the basis of Mr Scott's calculations which resulted in the eventual assessment. They argued that Mr Scott was inevitably influenced by the observations and calculations by Mr James; that by his own evidence Mr Scott had admitted that the assessment was not based on facts or evidence, but on a number of unverified assumptions (especially as to the uplift of 20% applied to the 10/01 quarter figures, which was then used as the basis for the remaining quarters), and accordingly the assessment must be regarded as arbitrary and therefore unreasonable; and that his approach of assuming that the levels of business in the first year would be the same as the levels of business in the second year was unreasonable as it ignored the reality that businesses usually struggle in their opening year.
43. They concluded by pointing out that the Appellant had given credible evidence, and had been prepared to concede that there had been some imperfection in his business record-keeping resulting in minor irregularities, and his offers during the investigation to settle a figure had been an honest attempt to quantify an under-declaration arising from imperfections of that nature.
44. For the Commissioners Mr Keller submitted that the test is whether the Commissioners' determination as appearing in the assessment was made to the best of their judgment, as the courts have applied that requirement. In his submission there were two stages to the analysis of whether the Commissioners had exercised best judgment: were they entitled to conclude that there was suppression of takings; if they were entitled so to conclude, were they then entitled to conclude that suppression had been at the levels indicated in Mr Scott's letter of 22 May 2002 and as subsequently applied in determining the assessment.
45. As to the first question, the Commissioners were entitled, on the evidence available, to conclude that there had been suppression of takings by the Appellant, even though it was accepted that the Appellant had not deliberately acted to under-declare his takings or the VAT due. The evidence provided by the surveillance by Mr James and his team was that sales which they had carefully observed and noted were not recorded in the records of takings; there was also the evidence that, on the 11 August 2000 visit the test and "overheard" purchases accounted for 88% of the declared takings for that evening, notwithstanding that there were long periods when officers were not observing purchases. Mr Keller was prepared to concede that some of the conclusions and assumptions used by Mr James could be questioned, but he pointed out that the Commissioners had not used these (or the resulting suppression ratios) to calculate the eventual assessment, but merely to reach the conclusion that takings had been under-recorded. The Appellant had been prepared to accept that there had been some level of under-declaration, and so the Commissioners' conclusion to that effect was not only reasonable, but conceded.
46. As to the second question, it must first be noted that the assessment (at least for the early periods) had to be made in the absence of any records of the business other than the Appellant's declaration of daily takings. Mr Scott therefore worked from later periods, taking into account declared takings for the period after the investigation was disclosed to the Appellant. He did not use the calculations of Mr James except as an indicator that there had been a significant level of suppression. For the busiest quarter (to 10/01) of the business he decided on an uplift of 20% which was reasonable in that it gave average weekly takings of £7,200, a figure below the "benchmark" of £8,000, below the declared takings of £8,503 for the week of 11 August 2001, and below the maximum output capacity of the business of £8,820 as subsequently calculated by the Appellant's accountants. Having concluded that an uplift of 20% was appropriate in giving a result consistent with the capabilities of the business, Mr Scott used that as the basis for calculating takings for earlier periods, factoring into the calculation an assumed 10% price increase to favour the Appellant. The methodology adopted by Mr Scott was reasonable, it was consistently applied, it was benchmarked against the likely capacity of the business, it reflected the seasonal cycle of the business and it was therefore reasonable to conclude that it gave a fair measure of the likely level of suppression. That being so, the Commissioners had assessed the amount of VAT due from the Appellant to the best of their judgment, in accordance with section 73(1) VATA.

The reasons for the Decision

47. Section 73(1) VATA provides:

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

48. It is our decision that the Commissioners were entitled to reach the view that the VAT returns of the Appellant were incomplete or incorrect, in that they under-declared supplies made by him and in consequence under-declared the amount of VAT due from him. It is also our decision that the assessment eventually made by the Commissioners on the Appellant was made to the best of their judgment, and accordingly should not be discharged. However, on the evidence before us, the material part of which became apparent only at the hearing of the Appellant's appeal, we consider that the assessment should be reduced on the basis we set out below.
49. It is well-established from the leading cases on assessments made to best judgment (see Van Boeckel v Customs & Excise Comrs [1981] STC 290 and Rahman (No 1) [1998] STC 826)) that an assessment made under the powers of Section 73(1) VATA can be set aside by the Tribunal only in the exceptional cases where the taxpayer establishes that it is wholly unreasonable, as where it is made capriciously or dishonestly, or where it is a "spurious estimate or guess in which all elements of judgment are missing". This is a high threshold for the taxpayer to cross, and we are clear that the conduct, approach and methodology adopted by the Commissioners in this case falls well within the bounds of what is reasonable or acceptable: in the circumstances where they had detected material suppression of takings they endeavoured, in the absence of any reliable records for certain relevant periods, to estimate the likely scale of suppression, and they made their estimate in the context of what they could judge as being the likely capacity of the business under review.
50. In the correspondence which followed the investigation, and also in the hearing of the appeal, the Appellant's representatives concentrated much of their energy on challenging the fairness and reliability of the observations made by Mr James and his team, and on the assumptions then made by Mr James when he calculated the initial suppression ratios first put to the Appellant. We share some of the concerns which they expressed: it seemed to us a serious flaw that, when the officers were instructed to adjust the hourly rota of visits so as to time them to coincide with the presence in the premises of other customers, it should then be assumed that the levels of business were the same when officers were not present as when they were – in other words, in the phraseology of Mr James, that business in the "time out" periods was at the same level as in the "time in" periods. It seemed to us that visits had been manipulated specifically to ensure that "time in" periods were busier than "time out" periods. It also seemed to us a serious flaw that (with only minor adjustment to reflect the very end of the trading period), the levels of business should be averaged out over the entire trading period, without any regard for the pattern of trading in the course of the evening.
51. Had the assessment been made on the basis of Mr James's calculations, then, having regard to these flaws, we would have had serious doubts as to whether they were made to best judgement. But what is clear is that the Commissioners themselves had reservations of some kind about the assumptions used by Mr James and the figures he produced, since the approach adopted by Mr Scott in calculating the figures used in the eventual assessment itself proceeded on quite a different footing and without recourse to the figures or assumptions used by Mr James. The Commissioners used the observation exercises by Mr James and his colleagues to satisfy themselves that there had been material suppression of takings by the Appellant, but they then disregarded the detailed analysis which Mr James made, including the flawed assumptions: as we have mentioned above, in Mr Scott's phrase in his evidence to the Tribunal, the observations and calculations made by Mr James opened their eyes to the fact of suppression, but did not provide empirical evidence of the likely amount of suppression.
52. We accept that the observation visits by Mr James and his colleagues demonstrated that there had been material suppression by the Appellant – not through any deliberate act on the Appellant's part, but by reason of poor procedures and record-keeping, and perhaps through lack of supervision of those working at the counter. It was established by those visits that a significant proportion of transactions which were carefully observed and noted did not appear on the till roll and were therefore not included in the declared takings for that evening. We note the point made by the Appellant that, in the case of the 8 February 2001 visit all of the actual purchases made by officers were shown as recorded, and that less reliance should be placed on "overheard" transactions; nevertheless, the detailed records of the officers concerned demonstrate that they were conscientious and careful in making their observations and in making a contemporaneous note of those observations, and that if they could not be certain of the price paid in any particular case, that "overheard" was disregarded when comparing transactions observed with transactions recorded on the till roll. Moreover, (and taking into account the factor already noted that officers adjusted the timing of their visits within the hourly rota they set themselves so as to observe as many customers as possible, so that they were likely to be present during busier periods), the fact that, in the case of the 11 August 2000 visit, officers observed transactions accounting for 88% of the declared takings that evening although they were present for less than 20% of the opening hours, gives further support to the Commissioners' view that there was material suppression (in the case of the 8 February 2001 visit, during a much quieter trading period, 66% of the declared takings were observed when officers were present for 23% of opening hours spread over the whole evening).
53. Although the focus of the Appellant's case was the calculations made by Mr James, that was misconceived on the Appellant's part. The essential question is whether the assessment, made on the basis of Mr Scott's quite different calculations, can be impugned on the grounds that it is so unreasonable as not to be made to best judgment. A preliminary issue here is whether the Commissioners, having decided that they should not use the observations made by Mr James and his team as the basis of the calculation (although having decided that in principle those observations showed suppression of takings to a material extent), should have tried to devise some other, more satisfactory, set of observations, or even have applied more sophisticated assumptions to the information gathered by Mr James. It seems to us that they were not obliged to do so. Of course, once they had disclosed their investigation to the Appellant (on 7 September 2001) there was no point in further observations, but it is perhaps strange that no observations were made between February and September 2001 by way of further verification. But the approach of the Commissioners was to confront the Appellant with their evidence of suppression and to seek his comments and any information he could provide from his business records so as to enable them to calculate a more accurate figure for the assessment. That seems to us to be an acceptable approach. It is not the responsibility of the Commissioners to monitor the taxpayer's business to establish that which the taxpayer himself is unable substantiate because of his inadequate records or other neglect. When the Appellant failed to provide any material information, in part because of his inadequate records, Mr Scott set about devising a method of calculation which he considered as being a reasonable basis for arriving at the amount under-declared.
54. The basis of Mr Scott's calculation which resulted in the eventual assessment was set out in his letter of 22 May 2002 to the Appellant's accountants. We had some difficulty in understanding from this letter exactly how Mr Scott had proceeded in his calculation until we had the benefit of having Mr Scott take us through it in his evidence at the hearing and also the benefit of the diagram he handed in at the hearing by way of further explanation. It is clear from the correspondence with the Appellant's accountants after the letter of 22 May 2002 that they likewise did not fully appreciate how Mr Scott had made his calculations, since there is little by way of challenge to his approach, but mostly further repetition of the objections to the figures produced at the earlier stage by Mr James. Perhaps they should have pressed Mr Scott for more explanation if at the time they were uncertain of his reasoning. In any event, it was only at the hearing that Mr Scott's assumptions and calculations were explained and then scrutinised.
55. Mr Scott's method of calculating the under-declared sales and VAT is set out at paragraph 35, to which reference should be made. In principle it appears to us that his method is reasonable and appropriate to the circumstances. It takes into account "reliable" figures from the business (those recorded after the investigation was disclosed in early September 2001), and relates those back to corresponding earlier periods, so far as possible, to form the basis of increasing the under-declared takings in those earlier periods. It is the case that the 20% uplift in the 10/01 figures (which is one of the key assumptions in the calculation) is to a degree a matter of judgment rather than calculation, but such a judgment is required in circumstances where facts are not available. Mr Scott exercised his judgment, and he tested it against such information as he had: the 20% uplift gave a figure for average weekly takings of £7,200 for the three months to the end of October 2001, and he noted that this was well within the declared takings for the week of 11 August 2001 of £8,503 (admittedly, this was likely to be a peak period for the business) and also the average weekly declared takings of approximately £8,000 for the weeks from the disclosure of the investigation to the end of October 2001. Although the information was not fully made available to him until after Mr Scott had made his calculations, his resulting figure of £7,200 can also be compared with the calculation made by the Appellant's accountants of £8,820 as the maximum output of the business, working from the cooking capacity of two chefs and an assistant chef. Therefore, although a case could be made for a different (higher as well as lower) uplift figure, Mr Scott, in choosing the figure of 20%, exercised reasonable judgment – it was not a "spurious estimate or guess in which all elements of judgment are missing".
56. We question, however, another key assumption underlying Mr Scott's calculations, and that is the assumption that the levels of business in the first year of trading were the same as in the corresponding quarters in the second year of trading (in the Appellant's favour Mr Scott assumed a 10% increase in prices in the second year – since Mr Ng told us that the Appellant did not in fact increase his prices, this assumption is tantamount to assuming that the volume of business increased by 10% in the second year). It will be recalled that Mr Scott used this assumption when relating back average weekly earnings established for the second year as the means of establishing average weekly earnings for the first year (see paragraph 35 (iv), (v) and (vi)). Mr Scott justified his approach by arguing that this was an established business known to the Appellant and which the Appellant took on as proprietor as a going concern, so that there was no basis for factoring into the calculation "start up" or other features associated with a new or regenerated business. We, however, heard evidence which leads us to conclude that this was too simple a view, and that there were factors which resulted in the first year of trading by the Appellant being less successful than the second year.
57. The Appellant's evidence, which we accept, is that during the first year (from opening in early June 2000 until the Easter holiday – mid April - in 2001) he was the only experienced chef working in the kitchen, assisted by Mrs Chung who cooked deep-fried dishes and also prepared any cold dishes (but cold dishes were not a very significant feature of the business). The Appellant was not able to cope with all the business at peak periods, and he lost custom as a result of the extended waiting times which ensued. He therefore turned to his family in London for help as he approached his second summer of trading, and from the Easter holiday in 2001, at all peak periods (week-ends, Bank Holidays and throughout the main summer holiday period) there were two experienced chefs at work in the kitchen, with Mrs Chung continuing as before.
58. We also note that, by the Commissioners' own report, the declared takings of the Appellant for the period from commencement of trade in early June 2000 to end of October 2000 were about 50% more than the takings declared by the previous trader (we assume for the comparable period 12 months earlier, but this is not clear). Mr Scott declined to comment on this when questioned, other than to say that as he understood it, the figures of the previous trader had not been verified in any way, and that he had not taken them into account in his assumptions as to the level of trading by the Appellant in his first year. Whilst we do not think that this report as to the previous trader's takings in itself forms a basis for recalculating the under-declaration by the Appellant, it does nevertheless give some credibility to the Appellant's general contention that it is too simplistic to assume a level pattern of trading.
59. In the light of the evidence we heard we conclude that there was a material factor which has not been taken into account in the assessment. We cannot blame Mr Scott for not taking into account what we might refer to as the understaffing of the kitchen (and the resulting problems for the business) in the first year of business since he was not made aware of that factor by the Appellant when the investigation was underway. However, it is a reasonable assumption that a new trader will experience some difficulties in the course of taking over even an established business, such that it is more likely than not that the second year of trading will show an improvement on the first year.
60. The fact that the assessment fails to take this factor into account does not mean that it should be set aside in its entirety on the grounds that it was not made to best judgment – any failure here was not a failure of judgment on the part of the Commissioners but, if anything, a failure on the part of the Appellant to bring the matter to the attention of the Commissioners when the matter was under discussion before the assessment was prepared. We also bear in mind the guidelines set out by the Court of Appeal in Rahman v Customs & Excise Comrs (No 2) [2003] STC 150 to the effect that in a "best judgment" appeal the principal responsibility of the Tribunal is to establish so far as possible the amount of tax properly payable.
61. We therefore decide that the assessment should not be discharged on the grounds that it is not made to the best of their judgment by the Commissioners, but we direct that the amount of the assessment should be reduced to take full account of the reduced cooking capacity (and consequent reduction in the turnover of the business) in the first year as against the full cooking capacity and consequent turnover of the business in the second year.
62. In the period from June 2000 to mid-April 2001 the business functioned with the capacity of one experienced chef and an assistant chef (Mrs Chung). From mid-April 2001 onwards at week-ends and holiday periods (including the peak summer holiday period) the business functioned with the capacity of two experienced chefs and an assistant chef. Therefore, in relating back to the period June 2000 to mid-April 2001 the takings calculated (as by Mr Scott) for the corresponding periods in 2001 and 2002 there should be a 30% reduction to reflect the reduced cooking capacity and resultant reduction in turnover capability in the June 2000 to mid-April 2001 period. This is in addition to the implicit reduction represented in Mr Scott's calculations by the assumed 10% price increase. We consider that a reduction of 30% is appropriate: on a straight arithmetic basis the reduction from two and a half chefs (so to speak) to one and a half chefs is a 40% reduction, but the increased capacity was not available throughout the second year (although it was available at the peak periods for the business), and no doubt there were times when the increased capacity was available, but the level of custom was such that it was not fully utilised. A discount on the reduction, from 40% to 30%, seems reasonable to reflect these factors.
63. We therefore direct that the assessment should be re-calculated by applying the principles set out in paragraph 62, with a consequent reduction in the interest payable, and that the assessment, as reduced by such recalculation, should stand. If the parties are unable to agree such a recalculation within the period ending 6 weeks from the date of this decision they are required to apply forthwith to the Tribunal, which will hear their respective arguments as to the correct calculation required to give effect to our decision and the Tribunal will then give a direction as to the amount of the revised assessment.
64. The Appellant's appeal is therefore dismissed, but the assessment shall be reduced as directed.
65. We make no order as to costs. We considered whether there should be some proportionate award of costs to the Appellant to reflect the reduction in the assessment, but since the matter which has resulted in that reduction was not raised before the hearing of the appeal we have concluded that the Commissioners should not be required to bear any part of the Appellant's costs.
A E SADLER CHAIRMAN RELEASED:21/06/2004

LON/03/314