British and Irish Legal Information Institute

Freely Available British and Irish Public Legal Information

[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 >> The Coffee Denn Ltd v Revenue & Customs [2009] UKVAT V20907 (06 January 2009)
URL: http://www.bailii.org/uk/cases/UKVAT/2009/V20907.html
Cite as: [2009] UKVAT V20907

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

The Coffee Denn Ltd v Revenue & Customs [2009] UKVAT V20907 (06 January 2009)

20907

Value Added Tax - Appeal against assessment based on the contention that turnover had been suppressed such that the business that had been de-registered as having a turnover below the registration threshold should have been re-registered when its adjusted turnover exceeded the threshold - Appeal allowed in part

LONDON TRIBUNAL CENTRE

THE COFFEE DENN LIMITED Appellant

and –

THE COMMISSIONERS FOR HER MAJESTY'S REVENUE & CUSTOMS Respondents

Tribunal: HOWARD M NOWLAN (Chairman)

MRS NORAH CLARKE

Sitting in Cardiff on 27 November 2008

Khan Rahman of Rahman & Co, for the Appellant

Mrs Crinnion of the Solicitor's Office of HMRC, for the Respondents

© CROWN COPYRIGHT 2008

DECISION

Introduction

We have found this to be a difficult case. The case revolved largely around trying to judge whether the Appellant had understated its turnover. The HMRC officer who had made assessments on the basis that the turnover had been understated, so that at some point the Appellant should have re-registered for VAT purposes, was undoubtedly right to observe that there were several very suspicious facts that suggested that turnover had been understated. The managing director of the Appellant was however impressive, and we concluded that the level of enhancement to the turnover was wrongly calculated by the HMRC officer. We were nevertheless unable to accept that the Appellant satisfied the burden of proof in asserting that its actual reported figures were accurate, and admitted of no adjustment. We have thus indicated a more modest basis on which we consider that the turnover figures should be adjusted and increased.

The appeal related to several years. In making our decision in principle, it was unnecessary for us to know the exact levels of reported and un-adjusted turnover in the various years, so that we cannot tell, at the point of releasing this decision, whether the revisions to turnover, on which we have decided, will render the Appellant registerable in the various VAT periods in dispute. It is nevertheless the case that if the adjustments to turnover (albeit fairly marginal) tip the Appellant over the registration thresholds for certain periods, the consequence for the Appellant will be that all the turnover in the affected periods will retrospectively attract liability to VAT, so that the consequence will be more serious than any consequence merely associated with the small increase in turnover that we have decided is appropriate.

The background facts

The Appellant company operated a coffee bar in Newton Road, Mumbles, Swansea. The two directors of the company were Mr. and Mrs. Dennison. Prior to the taking of the lease of the premises where the coffee shop was located the same company, under the same management, had earlier conducted a large newsagency business, employing 80 paper boys. Mr. Dennison, who gave evidence before us, had had a significant business career, having been a manager of an Asda store, if not an area of stores, prior to running the substantial newsagency business.

In 1999 Mr. and Mrs. Dennison realised that as Mr. Dennison was roughly 60 years of age, and Mrs. Dennison 54, and in particular as Mr. Dennison was suffering knee problems and arthritis (from which he was quite plainly suffering during the hearing), the newsagency business was too demanding. Mr. Dennison was a keen worker however and enjoyed cooking, so following the sale of the newsagency business, the Appellant company took a 15 year lease of the ground and first floor premises in a BT building in Mumbles, a fashionable and charming town.

Having spent a considerable amount of money in converting an empty shell into an attractive café, Mr. Dennison fairly soon came to realise that the new café venture was rather more problematic than he had hoped. Although it was located in Mumbles, it was apparently in a side shopping-street, and not in the main street that principally attracted the visitors. It was admittedly on the street that visitors would use if going to the castle, but apparently there was a more popular street, and a walk, neither of which passed the Appellant's café. It also transpired that most of the customers were elderly and particularly cautious of spending money. Thus Mr. Dennison came to realise that the café simply could not generate the required gross turnover to make a profit if he continued with the original business plan of employing a chef, and trying to provide a fairly full menu. In the period ending 30 June 1999 the company thus made a loss of £11,418 and had a wages bill of £23,706, and in the following year to 30 June 2000, the company made a much larger loss of £36,136 with a similar wages bill.

Faced with these losses and the conclusion that the turnover could not be increased to make a profit, Mr. Dennison looked at alternatives. He could not sell the business with its loss record, and a covenant in the lease never to cook with more than a domestic cooker (this restriction resulting from BT's concern in relation to fire when the telephone systems for the area of Wales were located above the café). He could not simply wind up the company because there was a 13-year residue of the lease, the rentals of which he had personally guaranteed. He thus evolved the plan of adopting a quite different business plan of cutting staff and costs, and abandoning the more ambitious menu with a view to trying to downsize and make a small profit with a much lower cost base.

A further important part of his plan was to eliminate the liability for VAT by ensuring that the turnover was always below the compulsory registration threshold. Since the company's rent, staff, and raw material costs did not attract VAT, the liability for VAT was a significant liability which he had effectively passed on to his customers in increased prices, so that he concluded that if he could keep the turnover below the relevant level this would compliment his down-sizing plan.

Whilst Mr. Dennison envisaged that the company's turnover would drop below the compulsory registration threshold once he had cut staff costs and the more ambitious menu, there were other ways in which he intended to ensure that he could control the company's turnover and keep it below the relevant level. He intended to close earlier than before, and to close occasionally during holidays, and he hoped to sub-let the property in order that others might effectively bear some of the rent burden, and then conduct their own businesses, distinct from the business of the Appellant, from the same premises. He also intended to cut down the heavy hours that he and his wife were working by generally not opening on Sundays. Whilst it may sound as if the week-end would be the best trading period in a place like Mumbles, it was suggested that because the Coffee Denn was located in a shopping street, rather than in the prime tourist area, its Sunday trading was likely to be very modest anyway, so that closing generally on Sundays would not be disadvantageous.

Pursuant to this revised business plan, the Appellant was de-registered for VAT purposes from 15 February 2001. Seemingly the new plan achieved its objective because in the periods ending June 2002 and 2003, there were modest profits of £4,371 and £2,997, with much more modest wage bills of £6,532 and £7,290 respectively. The Appellant traded in this manner and of course filed no VAT returns and had no further contact with HMRC in relation to VAT matters until the visit by the VAT officers, in the course of a "street sweep" on 12 May 2006.

The evidence

Evidence was given before us by Mr. Dennison, the principal director of the Appellant company; by Mr. Daniel McKay of HMRC, and on one minor point to which we shall refer later by Mrs. Sharon Hancox of HMRC.

We found Mr. Dennison to be a very impressive man. He had a slightly short fuse which we rather imagine had not served him particularly well in the few dealings that there had been in the past with HMRC officers, but he was clearly an experienced and competent businessman. In the past he had had, and had obviously had, more demanding jobs until his semi-retirement job of managing the Appellant company in its coffee shop venture. To a very large extent we were satisfied that Mr. Dennison was telling the truth, and we certainly did not consider that he had embarked on a plan of suppressing invoices in order to ensure that the Appellant's turnover was below the registration threshold. We did however hesitate to accept the totality of his evidence, or rather we consider that he failed to satisfy the modest burden of proof in relation to various matters. These matters, we considered, were fairly minor and marginal, as we will explain, though it may unfortunately be the result for the Appellant that this failure to convince us of the totality of his evidence may mean that the Appellant will be compulsorily registerable for certain periods, so that it is possible that the outcome might be almost as disadvantageous for the Appellant as if we had concluded that the bulk of Mr. Dennison's evidence was to be disbelieved. We will explain the points summarised above in due course and try to explain why we reach the somewhat strange conclusion of doubting just minor elements of Mr. Dennison's evidence.

We found Mr. Mckay's evidence to be entirely straight-forward, and we have not the slightest criticism of either his evidence or of the action that he took, and the arguments that he advanced. Mrs. Hancox gave evidence on one technical point that we will refer to below.

The course of the HMRC enquiries into the level of turnover of the Appellant

Mr. Mckay and a colleague made an unannounced visit to the Coffee Denn on 12 May 2006. The officers knew that the trader at the Coffee Denn was not registered, which they found a little surprising in view of the location of the business. The officers spoke to Mr. Dennison on the phone, since he was not then on the premises, and a meeting was arranged for 23 May. The only relevant point to note in relation to the first visit was that the officers noted that a Sunday menu was on display in the window.

At the meeting on 23 May, Mr. Dennison explained his strategy for ensuring that the Appellant's turnover was below the registration threshold, and said that as he was 66 he was certainly not inclined to do the further work that would be required to increase the turnover sufficiently to pay the VAT and still make a profit, which would all be required were the business to have a turnover above the VAT threshold.

At the meeting, the officers noted that the Sunday menu was not displayed, and that the opening hours displayed were from 10 a.m. until 5 p.m. six days a week. In discussion Mr. Dennison said that they only opened on Sundays on request, and that they had only opened on about 10 occasions on Sundays in the current year.

The officers took a Z reading from the till, and noted that the till record corresponded with the takings. They also asked Mr. Dennison how frequently he took Z readings, to which he replied that he took one daily, though of course there would be two on the day in question, since he had just taken a Z reading for the officers. Mr. Dennison explained that the till was a new one in relation to which they had initially had some teething problems, though it was now operating satisfactorily. At the meeting the officers examined the till rolls and Z readings for the period from 3 July 2005 to 14 May 2006, and noted to their surprise that the summer takings were not higher than those for the rest of the year. On leaving the premises, the officers took away the till rolls for the relevant 10 month period in order to examine them more carefully.

The till rolls that the officers took away had been produced by an earlier till, being an Olympia till. Mr. Dennison explained that this till had been very problematic, and that its instruction manual was almost impossible to understand. At the hearing, he explained the following matters about the operation of the Olympia till.

The method of operation of the till was that each item on the menu had to be programmed into the till, along with the information as to whether an order for that item should be transmitted to the kitchen or to the coffee area, and along with the price for the item. When thus soup and coffee were ordered, the orders would go to the kitchen and to a screen adjacent to the coffee machine, and the till roll would include the items bought and the price.

The Olympia machine could produce X readings and Z readings. X readings were a tally of all items ordered and charged since the last Z reading. X readings did not zero the till, so that if a subsequent X reading was taken after more orders had been rung through the till, or a Z reading was taken, the aggregate orders on the next reading would be those from the previous Z reading, i.e. including those both before and after the earlier X reading. Once a Z reading was taken, this Z reading would record all items rung through the till since the last Z reading, giving a total of takings since the last Z reading at the bottom of the Z print-out. The Z reading would then zero the till so that if one cup of coffee was rung up after the Z reading was taken, and a further Z reading was taken immediately, this would show only the charge for the cup of coffee, and the total for the Z reading at the same amount.

Z readings generally provided two other items of information. One was a sequential number for the particular reading along with a record of the date and time on which the reading was taken, and the other was an aggregate total of takings, running right through countless Z readings back to the date when the till was last re-set by the service personnel. The Olympia machine could apparently be programmed not to provide the sequential numbering of Z readings, but was not in fact programmed to omit the numbering.

When the officers examined the till rolls that they had taken away, they noticed two very significant things. They noted that the till rolls had been cut down so that whilst they still showed the total of takings for each particular Z reading, the information about the rolling aggregate total had been cut off and was thus not available for inspection. They also noticed that the print-outs that they had been given indicated, by their sequential numbering, that a great many print-outs were missing. For the period 3 July 2005 to 14 May 2006, the sequential numbering of the print-outs ran from 388 to 915, there thus implicitly having been 527 Z readings taken in the period, yet they only had print-outs for 211 Z readings. In the hearing there was a suggestion that they had print-outs for slightly more than 211 Z readings, but nothing much turns on this since the officers manifestly did not have print-outs for anything like the total number of Z readings that had in fact been taken in the period.

There was also something of a pattern in relation to the time when Z readings were taken, and the period when missing Z print-outs had been taken. Z readings were taken fairly regularly shortly after 5.00 p.m. on Monday to Saturday. The print-outs for these Z readings, albeit without their "tails" which indicated the total aggregate takings since the machine was last professionally re-set, were invariably in the bundle handed over by Mr. Dennison. Occasionally a second reading was taken on one of these days, whereupon that print-out was generally missing, but the vast majority of missing print-outs for Z readings had been taken on Sundays. In the bundle of print-outs, there was not a single print-out for a Sunday. Self-evidentially, since there were only about 40 Sundays in the period under review, and approximately 300 print-outs were missing, the sequential numbering on the adjacent Saturday and following Monday print-outs indicated that sometimes one Z reading had been taken on the Sunday, though quite frequently more were taken, and on one occasion 10 Z readings were taken on a Sunday, and none of them had been retained.

On being questioned by officers as to why the tails from the till print-outs had been cut off, Mr. Dennison said that they had been cut off to minimise paper and save space when storing the till rolls. When asked why some print-outs for Z readings were missing, Mr. McKay's written report suggests that Mr. Dennison appeared not to have realised that the print-outs were numbered. He said that on a Saturday, they would sometimes take a second Z reading, to ensure (as should be the case if no items had been rung up on the till since the last Z reading) that the previous Z reading had zero'd the till correctly, in order that in the following week the first Z reading would start from zero. Beyond this he would have to look at his records.

In due course, Mr. Mckay's concerns were seemingly confirmed by the two facts that phone records indicated that telephone calls had been made from the Coffee Denn premises on Sundays, and secondly demands had been made (sometimes on Sundays) to credit to the account of the Appellant the various credit card payments that the business had received.

Mr. McKay subsequently reviewed the records for earlier periods. Broadly the same pattern was exhibited by these earlier till print-outs.

Following his enquiries, Mr. Mckay calculated that the average daily takings in the period for which till print-outs had been provided had been £191. He then concluded that as there had been evidence of activity from the business premises on Sundays, and as Mr. Dennison had conceded at the earliest interview that the business had opened on about 10 Sundays in the relevant year, and that it opened on Sundays on request, and as numerous missing Z readings had been taken on Sundays (indeed virtually one or more on every Sunday, without exception), the Appellant must have been suppressing its turnover at the very least by destroying and ignoring the Sunday print-outs and thus by understating the turnover by that amount. He accordingly effectively grossed-up the turnover for all relevant years by the fraction 7/6 and concluded that in all periods after the business had been de-registered, it had in fact remained compulsorily registerable. He then made assessments on the re-computed turnover, and imposed the minimum 15% penalty for misdeclaration. To make the obvious point, the main significance of the adjustments was not that VAT merely became payable on the element of increase to the turnover, but naturally VAT became chargeable on the entire turnover.

On review, Mrs. Hancox reduced the assessment. This was for the technical reason that HMRC had lost the papers in reliance on which the Appellant had de-registered in early 2001. In view of this she accepted that HMRC are precluded from looking at any figures prior to the date of de-registration when considering whether and when the subsequent business turnover puts the trader above the registration threshold. Thus, ignoring the turnover before the de-registration, the threshold was only breached using Mr. McKay's adjusted figures (which she otherwise considered to be amply justified) in December 2001, so that, with the available one month's grace, the business became registerable on 1 February 2002. Beyond this, the consequence of the adjustment that Mr. McKay had made to the figures was that the adjusted takings resulted in the business being registerable in all later periods.

The relevant law

On the basis of the Respondents' contentions that the Appellant's turnover was under-stated, the Appellant has failed to make VAT returns for all periods since 1 February 2002, and the Appellant has also failed to keep all the records required to verify its own contentions as regards turnover. Under section 73 VAT Act 1994, HMRC may assess the Appellant the amount of VAT that they consider to be owing, and must do this to the best of their judgment.

On an appeal to this Tribunal, the Van Boeckel test, all as re-summarised by Mr. Justice Carnwath in Rathman (t/a Khayam Restaurant) v. C of C&E [1998] STC 826, indicates that we must follow a two-stage process in considering the appeal. Should we consider that the officer who made the assessments did not exercise "best judgment" which almost requires us to conclude that he acted "dishonestly, vindictively or capriciously", we should allow the appeal. If we consider that the officer did not fail in exercising his judgment on the first test, then we could nevertheless, on the second test, reduce the assessment if we considered on the evidence that we had heard that it was appropriate to reduce the assessment.

We should refer to two other legal points. First there is the non-contentious point addressed by Mrs. Hancox, referred to in paragraph 27 above. We consider that Mrs. Hancox made an appropriate adjustment in the light of the fact that all evidence of turnover in the period prior to de-registration had been lost and that the Commissioners had granted de-registration, and there is no need for us to refer further to this matter.

The other legal point to which we should refer becomes relevant because we have yet to summarise the facts that became clear during the hearing, namely that other people had traded from the Coffee Denn premises during the period under review. The relevant law in this context is that if as a legal matter, the right analysis is that all the trading was done by one trader, then naturally it is the full turnover of that one trader that is relevant in determining whether the compulsory registration threshold has been exceeded. Thus if we concluded that the "other trading" was rightly analysed to have been done by the Appellant, with the other ostensible separate trader in fact being an employee of the Appellant, or if we concluded that the right analysis was that the Appellant and the other ostensible trader or traders were in partnership then in either of these cases the aggregate turnover would all be attributable to one trader and the registration threshold would be applied accordingly. As a quite separate matter, HMRC are given powers to prevent the avoidance of VAT by the artificial separation of businesses that do not fall foul of the basic legal points just addressed. Under paragraph 1A to Schedule 1 to the VAT Act 1994, the Commissioners can issue a direction to treat the persons named in the direction as a single trader, where the Commissioners consider that this is appropriate in the light of the close links between the persons in financial, economic and organisational matters. No such direction has been given in this case, and since this power to treat separate persons as one trader only operates after the giving of the direction, it appears that this particular principle is irrelevant in this case, at least at present.

Counsel for the Respondents indicated that the present hearing was based entirely on the turnover of the Appellant alone, and that for the present HMRC had not considered any contention in relation to applying the registration threshold test by reference to the turnover of anyone other than the Appellant. We should comment that on reflection we have found this concession to be a little confusing. It seems reasonable to assume that the concession must have meant that it would not be argued that on general legal principles the other ostensible traders at the premises were properly to be regarded as employees of the Appellant or to be regarded as being in partnership with the Appellant, since, absent any issue of a direction under paragraph 1A just referred to, there could be no question of that second principle having had any application at all, regardless of any concession. On the other hand various facts emerged during the hearing in relation to the activities of two other ostensible traders, one being Mr. Dennison's son, the detail of which had clearly not been understood by the HMRC officers prior to the hearing. In the light of it not being absolutely clear that HMRC were advancing no argument before us that the trading of Mr. Dennison's son was to be treated legally as all undertaken by the Appellant, we will give some views in our decision on this matter.

The contentions on behalf of the Appellant

It was contended on behalf of the Appellant that:

• the fact that Mr. Dennison's son, Mark Dennison took a sub-lease of the premises and operated a, "bistro style" restaurant business, with a much fuller menu than that of the Coffee Denn, under the name "The Denn", from 4 May 2004 to 26 November 2005, operating on Wednesday, Thursday, Friday and Saturday evenings and on Sunday, precluded any argument that the Appellant was trading on Sundays in this period;

• the same fact accounted for why 98 of the till print-outs had not been retained by the Appellant, since the same till was used by both businesses, and the Z readings, and print-outs attributable to The Denn had been handed to Mark Dennison;

• Mr. Dennison had retained those parts of the till print-outs that were relevant for his purpose in drawing up accounts, particularly when he believed that the Appellant was not liable to VAT;

• had Mr. Dennison wanted to conceal the missing till print-outs he could have programmed the Olympia machine not to attach sequential numbers to each Z reading, but he did not do so;

• whilst at this distance in time, Mr. Dennison could often not account for why additional till readings had been taken, often but not always on a Sunday, he did suggest that the till had been difficult to operate, and a reading was often taken at the end of the week to check that the last Z reading had indeed zero'd the takings figure;

• an additional explanation of the additional Z readings was that it was often on Sundays that Mr. and Mrs. Dennison would amend the menus, and when they did this they had to programme each new or changed item and indicate the price for the item. In order then to check that the machine had recorded matters correctly, they would often operate the till as if a customer had ordered the new or changed item to see that the correct price came up, and would then have to take a Z reading to delete from the machine what would appear to be a sale, but was in fact simply a test order to check that the machine was operating properly;

• when the Olympia till was replaced in April 2006 by a more sophisticated till which did record and retain aggregate takings figures, HMRC officers had found nothing dubious or challengeable in relation to the figures produced on the new till, and there was relatively little difference between the Appellant's claimed turnover with the old till, and the undisputed figures generated by the new machine;

• when the business was closed on a Sunday, there would often be work to be done, cleaning and maintaining the premises, changing menu items, and preparing for the week ahead, so that it was entirely natural that the phone would be used on Sundays, without occasioning any reasonable inference that the business was trading;

• credit card slips were accumulated until there were sufficient to make it worthwhile transmitting them for collection. The credit card company rendered a charge for every occasion when payments were collected and credited to the Appellant's account, regardless of the number of payments claimed, so that it made sense to retain and collect several payments together and to have them all collected together. It followed of course that payments might well be collected on days other than the day when the items purchased had been ordered and paid for, and it was quite common for the collections to be made, when the business was closed, on Sundays;

• on the rare occasions when the business was open on a Sunday no Z reading would be taken on the Sunday so that the Monday Z reading would record the takings on both Sunday and Monday;

• Mr. Dennison's disability, and the fact that he had had knee operations in 2002 and 2003, aged about 63, amply justified the proposition that the Appellant would not generally trade on Sundays, particularly as business would generally not be good on Sundays when the adjacent shops were closed; so that

• the Respondents had failed to make out their case and to prove that the Appellant's turnover had been under-stated.

The contentions on behalf of the Respondents

It was contended on behalf of the Respondents that:

• the deliberate destruction of the tails of the till print-outs was highly suspicious, and it was not convincing for Mr. Dennison to say that he tore off the aggregate print out figures "to save space, when storing the till rolls";

• the fact that a Sunday menu was displayed when the officers first visited, coupled with the facts that Z readings were regularly taken on Sundays, the phone was used on Sundays and credit card payments were collected on Sundays as well as other days all suggested that The Coffee Denn was open on Sundays;

• Mr. Dennison had admitted that the business was open on Sundays on request, and that it had been open for about 10 Sundays in the year of the officers' visit but this was not matched by a single available till reading for a Sunday, and none of the Z totals for any Monday was notably higher than the average daily takings, so as to suggest that the Monday reading included takings for both Sunday and Monday; so that

• the Appellant had failed in proving, it having the burden of proof, that its declared figures were accurate, and that the increased turnover figures substituted by HMRC were inappropriate.

Our decision

We both confess that we have found this case difficult. Indeed we cannot say with any confidence what exactly happened, and whether and to what extent (if any) turnover was suppressed.

There were two periods of trading when the premises were used by some other operator. The first, in early 2001, lasted only a month and we now ignore it because it pre-dated the date on which HMRC now claim to treat the Appellant as being compulsorily registerable. We do however observe that rent of £1000 was apparently received from the other operator, and although the operator used the Appellant's till and credit card machine, it appears that all takings were accounted for to the third party. Accordingly it appears that the operator was neither an employee nor a partner of the Appellant.

We also accept that Mark Dennison, who was operating a more sophisticated bistro under a slightly different name from 4 May 2004 to 26 November 2005 was trading on his own account. He drew up separate accounts, and for two tax years filed Schedule D assessments on the very modest profits of his business. It thus appears to us that he was again neither an employee nor a partner of the Appellant. It follows that since all trading in the evenings and on Sundays in the period from 4 May 2004 to 26 November 2005 was undertaken by Mark Dennison, Sunday takings during this period were all reported in the books of the separate business.

We consider that Mr. McKay cannot be challenged for having been dishonest, vindictive or capricious in his enquires or in the estimates and assessment that he made. He was faced with the fact that more than half of the Z readings for the initial test period were missing, and that a somewhat absurd reason, namely saving on storage space, had been given for the reason why invariably the aggregate totals of Z readings had been torn off and were missing. Other evidence, such as the display of the Sunday menu on his first visit, the phone details and the credit charge collections plus the fact that many of the missing Z readings had been taken between the last Z reading on Saturday and the first taken on Monday evening suggested that the business had traded on Sundays. The basis of adjustment was then fair in that he simply grossed up the average reported takings for the six days by the fraction 7/6, and he disregarded occasions when more than one reading had been taken on other days.

Whilst thus we do not say that the assessment was made capriciously or that best judgment was not exercised, we do say that we consider that the evidence that emerged indicates that the actual method of adjustment was wrong, and that it should be modified. We might add incidentally that we imagine that Mr. McKay himself, on first understanding at the hearing that all Sunday and evening trading was undertaken by Mark Dennison and not the Appellant in the period from 4 May 2004 to 26 November 2005, would accept that it would no longer be appropriate to gross up the Appellant's turnover by reference to Sunday trading during this period.

We accordingly decide that no adjustment should be made to the Appellant's turnover for the period just indicated.

Our more material conclusion will appear to be self-contradictory and we will endeavour to explain why we consider that it is not.

We found Mr. Dennison to be an impressive businessman, and to be basically honest. We entirely accept that he devised his plan for down-sizing the business, trying to sub-let the premises to others to enable him to meet some of the rent bill without having to work all hours. We accept that his physical disabilities were very real and we have no doubt in concluding that he would have wished to cut down his working hours. We are also unable to accept that he embarked on a plan to under-state turnover fraudulently. We do not thus consider that the business was open on every Sunday and that there was a fraudulent plan to conceal approximately one-seventh of the takings.

Notwithstanding the conclusion in paragraph 42, there are too many oddities in the case advanced on behalf of the Appellant for us to accept it in its totality. The reason initially given for the removal of the tails from the till print-outs is ridiculous. It would have made more sense to say that aggregate totals were torn off because many of the missing till rolls were either zero Z readings, taken to check that the previous reading had zero'd the total, or better still to remove the confusion geared to test orders for new menu items being rung up to check that new items were recorded correctly. It is certainly fair to note that the totals would have included all these test readings, when in reality there was no turnover, so that in a sense, had this been said, the aggregate totals would have been misleading. Whilst that would have made sense, it was not asserted, and it is worth noting that the irrelevance of the test items would have been rather clearer if the missing print-outs had been retained. They would after all have presumably indicated that in a short period whilst the till was being programmed, several new items were rung up on the till to check that they had been correctly programmed.

In the Appellant's favour, it does seem obvious that many of the missing Z readings would not have been destroyed to conceal turnover. Occasionally there were as many as 10 Z readings taken between Saturday and Monday, and whilst one might have recorded Sunday takings, it is difficult to see why 10 Z readings would have been taken, other than because the Appellant's basic explanation about checking new menu items was true, albeit that it may not have accounted for every missing Z reading.

There are various reasons why, notwithstanding our conclusion in paragraph 42 above, we do not accept that the Appellant has fully proved its case and shown that its recorded turnover was precisely correct. The presence of the Sunday menu when the officers made their first visit, and the two admissions that the restaurant was open on Sundays on request, and that it had been open on about 10 occasions on Sundays in the year in question suggest that some trading definitely occurred on Sundays. There are occasions when Monday turnover was slightly higher than that on other days, but the same can be said of the turnover for every day in the week at some point, and we are certainly not convinced that Sunday turnover was always, or indeed ever, recorded instead on the Monday print-out. After all, why break the normal pattern of taking a Z reading immediately after closing, and why not take a Z reading on Sunday when in fact the missing print-outs, and Mr. Dennison's own evidence, indicate that many more Z readings were in fact taken on Sundays than on any other day of the week?

A rationalisation that seems to us to be possible is that Mr. Dennison did open the restaurant occasionally on Sundays, either on request, or perhaps to match expected demand. On the assumption that this was unusual, it seems to us possible that the till-rolls were then destroyed because of the annoyance and near offence that opening the café on a Sunday on an occasional basis would generate turnover that would destroy the basic VAT planning that was otherwise sound. There must be countless occasions where registered traders, say plumbers or decorators, might well report all their usual turnover entirely honestly for VAT purposes, but would nevertheless feel entitled to pocket cash if they did some week-end work for payment for a friend. Whilst that approach is of course wrong, there is a frame of mind that says that if a trader is prepared to work outside normal hours as a favour, he should at least be working entirely for himself and not for the Exchequer. In Mr. Dennison's case, the problem was of course far more serious because Sunday takings on isolated occasions would not just attract VAT but might render the entire turnover liable to VAT.

Our conclusion thus is that the Appellant did not embark on a plan to trade seven days a week, and to suppress till print-outs when they would tip the turnover above the VAT threshold. It is nevertheless possible that the Appellant traded occasionally on Sundays, and that this occasional Sunday trading was wrongly treated as being "outside the business". We do not say that this did occur, but we say that the burden of proof is on the Appellant (and not on the Respondents as the Appellant suggested) and the Appellant has not satisfied us that the occasional Sunday trading was, as the Appellant suggests, invariably included in Monday's print-outs. We have accordingly not been persuaded, on the balance of probabilities, that the Appellant's records were complete and accurate.

We thus suggest that the Appellant's turnover should be adjusted in a different way than the way in which HMRC initially adjusted it. As already mentioned, no adjustment should be made in the period between 4 May 2004 and 26 November 2005. Aside from this, it should then be assumed that the business was open on 12 Sundays in each year, other than for the ignored period, and that on those 12 Sundays, one in each month, the turnover should be increased by an amount equal to the average daily turnover for the year in question. To put it another way, the turnover should be grossed up by 12 rather than 52 Sundays.

In giving this decision we have already said that we cannot judge whether and when the adjusted business takings will exceed the registration threshold, and when the business will be liable to VAT, and when it will not be. Calculations will have to be done on the revised basis that we have suggested on a monthly basis. We were told that in two of the years the actual reported turnover was quite considerably below the registration threshold, and that means that it must be possible that the modest adjustment to turnover that we have decided is appropriate will still not tip the business over the registration threshold. And there will be bound to be a period within the period from May 2004 to November 2005 when the absence of any adjustment will mean that the business will not be registerable. Beyond this calculations will have to be done to assess the effect of this decision.

The 15% penalty imposed was the minimum penalty that can be imposed in relation to a misdirection, so that we make no change to that penalty, save in so far as the reduction of the assessment itself results in an automatic reduction in the penalty.

HOWARD M NOWLAN

CHAIRMAN

RELEASED: 6 January 2009

LON 2007/1061

About BAILII - FAQ - Copyright Policy - Disclaimers - Privacy Policy amended on 25/11/2010