England and Wales High Court (Chancery Division) Decisions

IN THE HIGH COURT OF JUSTICE
CHANCERY DIVISION
ON APPEAL FROM THE VAT AND DUTIES TRIBUNAL

		Royal Courts of Justice Strand, London, WC2A 2LL
		23 January 2006

Mr Justice Hart:

1. This is an appeal against a Decision of the VAT and Duties Tribunal (Lady Mitting, Mr Brian Strangward and Mr Robert Grice) made on 8^th March 2005.
2. The Tribunal treated as before it appeals against three assessments made by the Commissioners ("Customs") in respect of VAT in respect of the trade carried on by the appellant of Indian restaurateur at and from premises in Halesowen, West Midlands. The relevant notices of assessment were:

i) a notice dated 27^th April 1999 in the sum of £25,788 plus interest in respect of the trading period from 1^st August 1997 to 31^st December 1998 ("the first assessment");

ii) a notice dated 18^th May 1999 in the sum of £6,971.95 in respect of the trading period 1^st August 1996 to 31^st July 1997 ("the second assessment"); and

iii) a civil evasion penalty, raised under section 60(1) Value Added Taxes Act 1994, in the sum of £36,062 covering the period 1^st August 1996 to 31^st December 1998.

3. My reason for describing the Tribunal as having "treated" these as the assessments under appeal is to draw early attention to two complications about the background facts. One of those complications is that the notice of the first assessment in fact refers to a period from 1^st August 1996. This fact appears not to have been the subject of any debate before the Tribunal, it having been clear to all concerned that the notice in fact related to a period commencing on 1^st August 1997. The other complication is that in relation to the first period (i.e. that covered by the second assessment) Customs had, on 18^th June 1999, purported to amend it by giving notice that the assessment had been amended and increasing the sum assessed to £14,284. Customs subsequently had to concede that this amended assessment could not stand: there is no power under VATA for an existing assessment to be amended by increasing it. The figure of £14,284 continued however to play a part in the proceedings before the Tribunal since it was a component, together with the figures underlying the first assessment, of the calculation which underlay the civil penalty assessment. Both these complications are matters to which I return below (see paragraphs 41 and following below).
4. Putting those complications for the time being on one side, the potential issues on an appeal of this nature to the Tribunal were:

i) had the first assessment and/or the second assessment been made to the best judgement of the Customs?

ii) Whatever the answer to the first question, were those two assessments in the correct sums?

iii) Had Customs been justified in making a civil penalty assessment? For this purpose Customs had the burden before the Tribunal of showing that the appellant had dishonestly done or omitted to take some action "for the purpose of evading VAT": see section 60(1) VATA.

iv) If the answer to iii) is affirmative, what was "the amount of VAT evaded… or sought to be evaded" by the dishonest conduct? That amount is the measure of the penalty under section 60(1), subject to mitigation under section 70.

v) What mitigation of the penalty was appropriate?

5. The Tribunal's task in the instant case was eased by the fact that the appellant's representative before it expressly did not seek to contend that the first and second assessments had not been raised to the best judgement of Customs and, moreover, conceded that there had been dishonest suppression of turnover: see paragraph 44 of the Decision. The area of debate before the Tribunal was therefore over the second, fourth, and fifth of the issues which I have listed above.
6. Miss Lonsdale, who appeared before me on behalf of the appellant but who had not appeared below, relied on 7 grounds of appeal which, in her submissions to me, she advanced under three broad heads, namely:

i) unfairness in the proceedings before the Tribunal;

ii) the failure of the Tribunal to appreciate the true nature of its role in relation to determining the quantum of the assessments and the consequent failure of the Tribunal to address the issues raised by the evidence before it; and

iii) technical arguments in relation to the second assessment and the penalty assessment for the year to 31^st July 1997.

7. Before examining the arguments advanced under those heads, it is necessary to make some general observations about the Decision. It was made following four days of evidence and argument which took place on 10^th and 11^th May 2004 and 17^th and 18^th January 2005. The Decision itself is expressed in a document some 18 pages long, in single spaced typescript containing 65 paragraphs. Rather than attempting to summarise it, I can best introduce the factual issues which it addresses by quoting from the case which the Commissioners presented to the Tribunal in their formal Statement of Case (served pursuant to Regulation 8 of the VAT Tribunals Rules 1986). Analysis of the Decision demonstrates that the Tribunal found the allegations in the case to be proved. Those allegations were:

"1.Mr Liaquat Ali (hereinafter referred to as "the Appellant") carries on business as the sole proprietor of an Indian restaurant, catering for both eat-in and takeaway meals, know as "Vakas Balti", from premises at 64 Windmill Hill, Colleygate, Halesowen, West Midlands B63 2BZ ("the premises").

2. The Appellant is now registered for the purposes of value added tax under registration number 695 2458 93 with effect from 1^st August 1996.

3. On 2^nd July 1997, an Officer of Customs and Excise visited the premises and established that the Appellant was not registered for VAT.

4. During a further visit by the Officer on 1^st August 1997, the Appellant informed her that he was not trading above the VAT threshold but intended to register for VAT. The Officer advised him that his level of trading would need to be established and issued him with a set of invigilation sheets, which the Appellant agreed to complete for the month of August 1997. These sheets were completed on a nightly basis with the Officer making unannounced visits to check on their completion.

5. In addition to the said Officer's visits, a number of Officers of Customs and Excise carried out test purchases at the Appellant's restaurant and also undertook observations.

6. The Commissioners discovered that the self-invigilation sheets failed to record all of the Officers' test purchases and that the Appellant's business records bore no resemblance to the number of customers observed.

7. The Appellant finally registered for VAT with an effective date of 1^st August 1997. However, the application for registration stated that the threshold was exceeded on 1^st August 1996 and, therefore, following further checks by the Commissioners, the effective date was amended to 1^st August 1996.

8. On 4^th March 1998, Officers of Customs and Excise visited the premises once again to discuss the Appellant's business activities and to examine the available records.

9. It was established that between 7^th September 1997 and 21^st December 1997, the Appellant had purchased a total of 150 order pads (containing 50 orders each). When the Commissioners analysed the invigilation sheets they discovered that the total orders stated to have been used between 1^st August 1997 and 31^st December 1997 equated to 34 pads; there was therefore a difference of 116. This represented a suppression rate of 77%.

10.On 15^th May 1998, Officers carried out external observations at the premises, prior to making an unannounced 'walk in' visit to check takings and uplifting meal bills.

11.At approximately 01.30 hours on 16^th May 1998, the said Officers counted the takings, in the presence of the Appellant, and ascertained that they consisted of £1,030 in notes and coins, including a £75 float, £65.65 in cheques (3 cheques in total) and £264.45 in credit/debit card transactions (8 such transactions in total). The total takings for the night (less the declared float) therefore came to £1,267.10. This was £820.05 more than the single highest ever declared takings on a Friday night since the original date of registration on 1^st August 1997 and more than £1,033.30 higher than the average declared Friday takings between 1^st August 1997 and 14^th March 1998.

12.On the basis of these findings, the case was referred to the Commissioners' Local Fraud Unit and adopted for investigation.

13.On 20^th May 1998, Officers of Customs and Excise attended the premises and interviewed the Appellant in the presence of his accountant Mr Amir H Raja. During the course of the interview, the Appellant admitted that the meal bills observed by the Officers on the night of 15^th/16^th May 1998 were for meals sold that night. The said meal bills alone confirmed that of the £1,267 takings observed at least £912.50 must have related to takings for that night. No other admissions were made in respect of evasion of VAT and very little co-operation was given.

14.Officers made further observations at the premises on 10^th and 23^rd March 1999.

15.On 24^th March 1999, Officers once again attended the premises and conducted another interview with the Appellant, in the presence of his accountant. During the course of the interview the Appellant admitted that he had cooked more than two takeaway meals the night before despite having only declared two takeaway meal bills (Officers had in fact observed eight takeaway sales that night). He also admitted that he thought that an employee had destroyed some of the meal bills that night. He accepted that he had been aware of his VAT responsibilities. The Appellant gave little co-operation, however, and continued to deny suppression.

16. The Officers then informed the Appellant that they considered there to be sufficient evidence to issue an assessment. The Appellant stated that if the Commissioners sent him a reasonable estimate he would pay it.

17.On 27^th April 1999, the Commissioners duly issued the Appellant with a Notice of Assessment for an amount of £25,788 (plus interest). In addition, a manual Notice of Assessment was issued on 18^th May 1999, for £6,791.95; this assessment was subsequently amended on 18^th June 1999 to £14,284.47 (this assessment was for the period 1^st August 1996 to 31^st July 1997 and was issued on VAT 152A because the Appellant failed to render the VAT 100 for this period, which he was issued with on 15^th January 1998). The total arrears were therefore calculated to be £40,072 (based on a suppression rate of between 50% and 75%).

18.On reviewing all of the evidence the Commissioners also concluded that the conduct, giving rise to the net under-declaration of £40,072, involved dishonesty and on 2^nd November 1999 they made an assessment under section 76 of the Act to a penalty under section 60(1). In recognition of the limited assistance given by the Appellant, the penalty was mitigated by 10% of the culpable arrears, to £36,062.

19.Subsequent to the issue of the assessments, the Appellant's accountants sought a local reconsideration on the grounds that they had undertaken an analysis of purchases by his client and, as a result, had calculated that the true arrears were £8,012.63.

20.The Commissioners duly undertook reconsideration and concluded that the accountant's calculations, based on supplier records, did not provide a reasonable basis on which to make a calculation. As a result the assessments were upheld in full.

21.The Commissioners rely on the following facts to prove dishonesty in this matter:

(a) The quantity of cash found in the till on the night of 15^th /16^th May 1998 was far in excess of the takings recorded for any other Friday evening. As pleaded above, Officers ascertained that the total takings for the night came to £1,267.10, which was £820.05 more than the single highest declared taking on a Friday night since registration. Furthermore, £1,267.10 was more than £1,033.30 higher than the average declared Friday takings between 1^st August 1997 and 14^th March 1998. The quantity of meal bills found on that night substantiates the fact that the amounts did represent takings and the Appellant confirmed in interview on 29^th May 1998 that the meal bills observed by Officers were for meals sold that night.

(b) There are substantial discrepancies between meals recorded on the self-invigilation sheets and those seen by Officers who attended the restaurant between 1^st August 1997 and 4^th September 1997. Furthermore, the invigilation sheets failed to record all of the Officers' test purchases.

(c) Between 7^th September 1997 and 21^st December 1997 the Appellant purchased 150 order pads, whilst an analysis of the self-invigilation sheets suggests that the total orders used between 1^st August 1997 and 31^st December 1997 equated to 34 pads. In other words there was a difference of 116.

(d) The recorded credit card sales for some nights were in excess of the declared sales figures.

(e) In interview on 24^th March 1999, the Appellant admitted that he had cooked more than two takeaway meals the night before, although only two takeaway meal bills were declared. The Appellant therefore knew that the declared figure was understated. The Appellant also stated that he thought that one of his employees had destroyed some of the meal bills that night. This is further evidence that he knew that the remaining meal bills did not equate with the number of meals actually sold.

(f) The Appellant was well aware of his obligations in relation to VAT (as he admitted in interview on 24^th March 1999).

22.The Appellant's case, as set out in his Notice of Appeal dated 16^th June 2000, is that:

"The assessment is based on estimated [figures], which are not realistic or achievable by the business of this size and nature. We have done a lot of work to calculate a reasonable figure to any output tax underdeclared due to Mr Ali's illiteracy or ignorance but HMCE is not willing to review their decision".

23. The Commissioners contend that:

(a) There is clear evidence in support of the assessments (the Appellant is referred to the matters pleaded in paragraph 21(a) to (d) above) and in support of the calculations and figures which underlie them. The Commissioners calculated a suppression rate of between 50% and 75% based on the average number of meal pads calculated to have been suppressed and on the takings recorded on 15^th/16^th May 1998 as compared to takings on other Fridays. Furthermore, the figures are perfectly realistic and achievable by a business of the size and nature of the Appellant's.

(b) In so far as it is alleged that the assessment was not issued using best judgment, there was sufficient and adequate information upon which to make such judgment and the Commissioners considered all of the relevant information before issuing the assessment. There is nothing to suggest that the Commissioners have not exercised best judgment in carrying out the assessment. The Appellant is required to show that the assessment was made either dishonestly, vindictively, or capriciously, or is a spurious guess or is wholly unreasonable in the circumstances. The Appellant has provided no evidence in support of such allegations and any such allegation is manifestly ill founded.

(c) In the circumstances, having regard to the matters pleaded in paragraph 21, the civil evasion penalty has been properly and lawfully imposed on the Appellant. Furthermore, the amount of mitigation allowed to the Appellant pursuant to section 70 of the Act is reasonable in light of the Appellant's very limited co-operation.

(d) In the premises, the Commissioners contend that the appeal should be dismissed."

Alleged procedural unfairness

8. Miss Lonsdale identified eight shortcomings in the procedure adopted at the Tribunal hearing which, she submitted, taken together vitiated the Decision. I shall deal with them in the order in which Miss Lonsdale advanced them in her written skeleton submissions, but record the fact that in her oral submissions she placed the sixth criticism at the forefront of her argument.
9. First, it was submitted that correspondence which had taken place between Customs and one set of the appellant's representatives prior to the hearing had been opaque. The thrust of this submission was that the correspondence in question betrayed a failure by Customs, in the course of its own review of the assessment, to understand points being put to it on behalf of the appellant. The chief example cited of this was a letter dated 16^th March 2000 where there does indeed appear to be some lack of precision in distinguishing between numbers of diners observed by Customs and the number of bills declared.
10. This point was said to go to the Tribunal's view of the degree of co-operation given by the appellant to Customs, and therefore to the question whether the penalty had been adequately mitigated by the 10% reduction applied by Customs and upheld by the Tribunal.
11. In my judgment there is nothing in this point so far as the appeal before me is concerned. To the extent that there was ever anything in it it was a point available to the appellant at the Tribunal, and one whose determination was essentially a question of fact. I was quite unpersuaded that the appellant's ability now, through his counsel, to identify an apparent mistake in the letter concerned demonstrated some failure by Customs to provide the appellant with an opportunity to afford Customs the co-operation which they sought. As Mr Puzey (who appeared before me on behalf of Customs) pointed out, the letter has to be seen in the full context of correspondence which took place between 8^th July 1998 and 27^th April 2000, and of the interviews and meetings between the appellant or his representatives on 20^th May 1998, 24^th March 1999, 9^th June 1999 and 4^th September 2000.
12. Secondly, reliance was placed on the fact that the Customs officer whose name was subscribed to the assessment made internally on 16^th April 1999 was a Mrs Drew, not the Mr Townsend who had (basing himself on figures arrived at by Mrs Drew) made the final calculations.
13. I found it difficult to see exactly what point was being made under this head. Both Mr Townsend and Mrs Drew were tendered as witnesses before the Tribunal which noted, at paragraph 31 of the Decision, the nature of the roles each had played. Moreover even if (which is far from clear) those roles only became apparent to the appellant or his representatives at the hearing itself, the course which the hearing took in practice allowed the appellant and his representatives ample opportunity to make what they could of it in cross-examination of Mrs Drew and Mr Townsend, and in argument.
14. Thirdly, it was submitted that additions to the bundle of documents were made shortly prior to the hearing and suggested that these late additions unfairly prejudiced the appellant. This was not, however, a matter of which complaint was made at the time by the appellant's representative, nor was Miss Lonsdale able to persuade me that it was a matter of which complaint could legitimately have been made. The additions included witness statements which had already been served on the appellant and not objected to and schedules from Customs summarising information and evidence already held by the appellant. To the extent to which the additional material might have caused the appellant difficulty, or provoked a request for yet further disclosure, ample opportunity was again provided to the appellant and his representative to raise those matters given the course which the hearing in fact took.
15. Fourthly, objection was taken to the fact that, although seven witness statements were served by Customs, none of the witnesses exhibited or referred to the calculations which had been made by Customs. I did not understand this objection. The witness statements in question had been served prior to the hearing (and had not been objected to) without provoking any request for the disclosure which it is now said should have accompanied them.
16. Fifthly, it was objected that Mr Townsend had provided no witness statement prior to the hearing, and that when he did give evidence he produced no calculations of how he had arrived at the suppression rates of 50% which he had applied in raising the assessments. Once again, however, I am unable to see how this can be said to have amounted to procedural unfairness. Mr Townsend's approach to the figures, and the basis of his choice of suppression rates was, right or wrong, clear and is recorded in paragraphs 33, 34 and 35 of the Decision.
17. Sixthly, it was objected that Customs had failed to disclose the external observations which had been made on 15^th May 1998 and which are referred to in paragraph 10 of Customs' Statement of Case. This was a point which had, it appears, been noticed by no one until spotted by Miss Lonsdale shortly before the hearing before me. She rightly pointed out that evidence in relation to the unannounced "walk-in" visit which took place on that Friday night was of great importance to Customs' approach to the case: see paragraph 11 of the Statement of Case. The takings for that night identified by the officers concerned amounted to £1,267.10, which was £820.05 more than the single highest ever declared takings for a Friday night and more than £1,033.30 higher than the average declared for a Friday night. As is apparent from the detailed review of the evidence in relation to this episode contained in paragraphs 22 to 25 of the Decision and the Tribunal's conclusion at paragraphs 49 to 51, the Tribunal rejected the various explanations which Mr Ali gave for this discrepancy. Miss Lonsdale's submission was that the failure of Customs to disclose a record of the external observations gave rise to the inference that the product of those external observations must have been inconsistent with the case which Customs wished to advance.
18. This submission implied that the failure to disclose was part of a deliberate policy of suppression on the part of Customs with a view to misleading the Tribunal. Such a suggestion cannot in my judgment be supported. In the time available those instructing Mr Puzey were unable to produce any document evidencing the external observations which are referred to in paragraph 10 of the pleading, but Mr Puzey was able to tell me that such a document did not appear to have been included in the papers which were before counsel who had settled the pleading. In fact the only evidence before the Tribunal, or me, that any such observations were made prior to the "walk-in" visit is that contained in the pleading itself. Both the officers who made whatever observations were made on 15^th May 1998 (Mr Barlow and Mrs Drew) gave evidence before the Tribunal but neither was asked about the observations. The suggestion that there was any deliberate policy of suppressing a document is belied by the fact that the pleading itself refers to the observations.
19. Seventhly, Miss Lonsdale reiterated her criticisms as to the lack of disclosure accompanying the unchallenged witness statements. I was unable to see how this objection differed from her fourth objection. As it seemed to me the point Miss Lonsdale was really making was that, if it had fallen to her to conduct the case below on the appellant's behalf, matters would have been dealt with in a very different way. I do not doubt that, but cannot see that it provides grounds for a successful appeal.
20. Eighthly, it was submitted that the suddenness of the withdrawal at the hearing of the assessment dated 18^th June 1999 (see paragraph 3 above) deprived the appellant of the opportunity of considering the legal consequences of the withdrawal. This submission is more conveniently addressed in the context of the technical arguments raised by Miss Lonsdale in relation to that assessment which I consider later in this judgment.

The failure of the Tribunal to appreciate or properly to execute its true role

21. The submissions under this head fell into two parts. First it was said that the Tribunal had failed to distinguish between its supervisory role (in determining whether the assessment had been made to best judgment by Customs) and its original role (in determining the correct amount of the relevant assessments); and, secondly, that insofar as it had sought to determine the correct amount of the assessments, it had reached a conclusion at which no reasonable Tribunal could have arrived on the evidence before it.
22. The distinction between the Tribunal's supervisory and original roles in relation to best judgement assessments is illustrated and discussed in a number of decisions: see, in particular, Rahman v. Customs & Excise Commissioners [1998] STC 826 at pp 836c-f, 839b-d (Carnwath J.), Rahman v. Customs & Excise Commissioners (No. 2) [2002] EWCA Civ 1881, [2003] STC 50, at paragraphs 6 and 40 (Chadwick LJ., with whom Brooke LJ. and Bodey J. agreed), and Customs & Excise Commissioners v. Pegasus Birds [2004] EWCA Civ 1015, [2004] STC 1509 at paragraph 38 (Carnwath LJ.) and paragraphs 87 to 92 (Chadwick LJ.).
23. The suggestion that the Tribunal was unaware of the distinction between its supervisory and original roles seems to me to be unsustainable for two reasons. First, it plainly had in mind the distinction when, at an adjourned hearing of the appeal on 15^th September 2004, it drew to the attention of the appellant the recent decision in Pegasus Foods, inviting him to consider whether he wished to make "amendments to his pleaded case in so far as it relates to best judgement". The Tribunal's reference to the case is incompatible with the suggestion that it did not understand the fundamental message of that authority, namely that:

"the tribunal should remember that its primary task is to find the correct amount of tax, so far as possible on the material available to it, the burden resting on the taxpayer. In all but very exceptional cases, that should be the focus of the hearing, and the tribunal should not allow it to be diverted into an attack on the Commissioners' exercise of judgement at the time of the assessment." [see para. 38(i) ibid].

It seems likely that what the Tribunal had in mind in drawing the authority to the attention of the appellant was that, if he intended to challenge the assessment as a whole on best of their judgement grounds:

"it is essential that the grounds are clearly and fully stated before the hearing begins." [para. 38(ii) ibid]

The Tribunal seems, therefore, to have been seeking to elucidate at that stage whether it was going to be asked to consider "best judgement grounds" separately from the question as to what was the correct amount of tax. Not only does the Tribunal's own reference to Pegasus Foods make it clear that it perfectly understood what its role was, but its reference to Pugh in paragraph 57 of the Decision also demonstrates that it was entirely at ease with the distinction between "best judgement" considerations and issues of quantum.

24. Secondly, the appellant's representative at the Tribunal hearing made it clear that the assessments were not being challenged on best judgement grounds. The Tribunal recorded this both at paragraph 44 and at paragraph 48 of its Decision:

"Mr Khan… did not seek to contend the assessment had not been raised to best judgement but that it was grossly excessive"

and:

"Mr Khan accepted the assessment was raised to best judgement and his challenge to it was focused on quantum – both as to the level of suppression assumed by the Commissioners and the period over which the Commissioners assumed the suppression had been taking place."

It seems to me impossible to say that the Tribunal did not understand that the issue before it was the issue of quantum.

25. It was indeed the case that, in considering the rival cases being advanced before it, the Tribunal necessarily had to consider whether the assumptions which had been made by the Commissioners were reasonable ones on the evidence before it. That was because the Commissioners were relying on the same evidence before the Tribunal as they had themselves relied on in making the assessments. Once the Tribunal had rejected the alternative figures put forward on the appellant's behalf, it was inevitable that they should consider whether the evidence was capable of supporting the inferences which the Commissioners had drawn from the evidence in making the assessments and were now inviting the Tribunal to accept as correct.
26. I turn then to Miss Lonsdale's submission that the Tribunal, in assessing the evidence before it, reached a conclusion at which no reasonable tribunal could have arrived. This involved her in making detailed submissions on the effect of the evidence. For the most part those appeared to me to be no more than a re-casting, by reference to new schedules prepared by her on a variety of different assumptions, of points of a kind which the appellant's representative had made on his behalf at the Tribunal. In one respect, however, Miss Lonsdale did focus on an issue which, she argued, revealed a logical flaw at the heart of the Tribunal's reasoning.
27. The Tribunal records at paragraph 33 of the Decision that in raising the assessments Customs, having concluded that there had been some suppression of sales (i.e. that actual takings had been in excess of declared takings), had asked themselves what the appropriate suppression rate was. The evidence they had from their visit on 15^th May 1998 indicated a suppression rate of 80 per cent. This tallied with a figure of 77% produced on the basis of the number of order pads produced. Customs rounded that figure down to 70% and applied that to each of the VAT periods from 1^st August 1997 to the end of March 1998. They observed, however, that while declared takings appeared to have been reasonably constant during 1996 and 1997 declared takings increased in the first three months in 1998 and again in the period to 30^th June 1998, giving an increase over this period of 33 per cent. Taking the view that this increase reflected a more honest declaration, Mr Townsend reduced the suppression rate for the periods 6/98, 9/98 and 12/98 to 50 per cent. Thus the first assessment was made on the basis of a 70% suppression rate for the period from 1^st August 1997 to 31^st March 1998 and a 50% rate for the remainder of 1998.
28. Miss Lonsdale submitted that the Tribunal cannot have fully appreciated the practical effect of using these suppression rates. As she pointed out, a 50 per cent suppression rate implies that for every 30 meals declared the appellant was not declaring 30 meals, whereas a 70 per cent suppression rate means that for every 30 meals declared he was suppressing 70 meals. The implications for turnover are equally striking. Using a 50 per cent suppression rate a declared turnover of £100.00 implies an actual turnover of £200.00. Using a 70 per cent suppression rate a declared turnover of £100.00 implies an actual turnover of £333.33.
29. This does, in my judgment, cast some doubt on the process of reasoning which led Customs to reduce the suppression rate in the way it did in response to the increase in the declared takings in the first half of 1998. That increase was said to have been by 33% and was ascribed by Customs to a greater honesty on the part of the appellant. However the arithmetic of their assumptions produces a strange result. If declared turnover in 1996 and 1997 is taken as 100 in each quarter, then the actual turnover during each of the periods to which the 70% rate was applied will have been 333. But when the declared turnover increases by 33% (to 133), the 50% rate implies that actual turnover must have dropped to 266 (whereas a 70% suppression rate would indicate an increase in actual turnover to 443).
30. There was no evidence before the Tribunal that the appellant's actual turnover had decreased so significantly between the first six months and the second six months of 1998. On the contrary there were two reasons to suppose that actual takings in 1998 are likely to have been higher than in 1997. The first is the fact that, as recorded by the Tribunal, declared takings rose. The second is that, at some point in late 1997 or in early 1998 a store-room in the premises had been converted to restaurant use, thus increasing the maximum number of covers from 68 to 108.
31. The conclusion to which this drives me is that the 70% and 50% suppression rates used by Customs, and accepted by the Tribunal, cannot both have been right. However, it seems clear that, of the two figures, it is the 50% that has no support in the evidence. The 70 % figure had a firm evidential base, being based on a comparison between the actual takings (£1,267) on the night of Friday 15^th May 1998 and comparing them both with the previous highest declared takings for a Friday night (£447, implying a 65% suppression rate), and the average Friday night takings between 1.8.97 and 14.3.98 (£234 implying an 81.5% suppression rate), and supported by the order pad evidence (which implied a suppression rate of 77%).
32. Once one factors in the potential effect of the conversion of the store room to restaurant use, it is possible to criticise the use of an average taken from the whole period from 1^st August 1997 to 14^th March 1998, especially if (as Miss Lonsdale urged me to suppose, although there was no direct evidence on the point) the storeroom conversion had only taken place in March 1998. What one needs to do is to make allowance for the fact that that average reflects the lower pre-conversion turnover. On the evidence before the Tribunal one possible approach would have been to assume that that increase in trade which resulted from the conversion corresponded proportionally with the increase in declared takings observed by the end of June 1998 (an increase of 33%). Then, if one takes the 15^th May 1998 to have been a typical post-conversion Friday night, that evening's takings have to be reduced by one quarter in order to find a comparable pre-conversion figure. That gives a figure of £950 hypothetical actual takings for a pre-conversion Friday night to be compared with the average declared figure of some £234. But that still yields a suppression rate of 75%.
33. Miss Lonsdale criticised the Tribunal for not having investigated the date of the conversion, or come to some conclusions as to its effect. I do not think that that criticism is justified. While it was the function of the Tribunal to arrive at a conclusion as to the correct figure, it could only properly do so in the light of the arguments and evidence presented to it. Its role is not inquisitorial. It is perhaps understandable that the appellant's representative at the Tribunal did not choose to direct its attention to these matters. As the calculations in paragraph 32 above demonstrate, to have done so might have led into a forensic trap. The conclusion might well have been that Customs had been wrong to suppose that the increase in declared takings was the result of a greater honesty on the part of the appellant: the real explanation was that actual takings had increased. If there was no reason to suppose a greater honesty, there was nothing to support the belief that a 50% suppression rate was appropriate in the latter half of the year. It is hardly surprising that the view was taken that the appellant's best chance of upsetting the assessments was to attack the essential premise on which they were based, namely that the takings for 15^th May were not £1,267. For the reasons given in paragraphs 49 to 51 of the Decision the Tribunal, however, rejected the appellant's alternative explanations for that figure.
34. For those reasons the Tribunal was in my judgment entitled to conclude that the approach taken by Customs was a fair one to adopt, in the sense that it did not lead to an assessment which was in excess of what would have been due from the appellant had he made honest returns. Once one accepts that the takings for 15^th May 1998 were £1,267, an assumed suppression rate of 70% is perfectly rational. The application of a 50% rate to the latter half of 1998 and to the year to 1^st August 1997 has less to be said for it, but is not a matter of which the appellant can sensibly complain given the Tribunal's rejection of his alternative approach to the calculation.

Technical Arguments

35. The arguments under this head all related to the validity of the assessments raised for the period from 1^st August 1996 to 31^st July 1997. In considering them the following chronology has to be kept in mind:

i) Customs initially required returns to be made for the periods 1^st August 1997 to 30^th September 1997 and for 1^st October 1997 to 31^st December 1997;

ii) Subsequently, on 15^th January 1998, Customs required a return to be made for the period 1^st August 1996 to 31^st July 1997. The figures supplied to Customs for that period showed VAT due of £6,971.95. That sum was then assessed by the second assessment (made on 18^th May 1999);

iii) The first assessment (in respect of the period from 1^st August 1997 to 31^st December 1998) had been made on 27^th April 1999. This was the assessment which proceeded on the assumption of the 70/50% suppression rate;

iv) After the first and second assessments had been made Customs re-visited the question of whether there had also been suppression in the 12 month period ending on 31^st July 1997, and concluded that there was likely to have been such suppression. They used a 50% rate to calculate the amount of suppression, and on that basis concluded that the VAT due in respect of that period should have been £14,284 rather than the £6,971.95 which had been declared. They therefore purported on 18^th June 1999 to amend the second assessment by the amended notice referred to at paragraph 3 of this judgment. Since there is no power to amend an assessment by increasing the amount payable, this purported amended assessment could not be relied on by Customs before the Tribunal. The correct procedure for Customs to have taken would have been to issue a further or supplementary assessment or to have withdrawn the earlier assessment and replaced it: see sections 73(6), 75(2) and 77(6) VATA. By the date when the mistake was realised, it was too late to cure it;

v) The civil penalty assessment was calculated on the basis that the figure contained in the "amended" assessment dated 18^th June 1999 correctly reflected the amount of VAT which had been evaded in the period which it purported to cover.

36. Against that background Miss Lonsdale advanced four arguments. I will take them not in order in which she made them, but ranking them according to their potential consequences.
37. The first argument is that Customs had no power to issue the requirement on 15^th January 1998 for a global return to be made for the 12 month period ending on 31^st July 1997. If that was correct it followed that there had been no prescribed period for making a VAT return. The consequence was that the second assessment was bad, and the civil penalty, insofar as it was based on the amount of VAT evaded during that period, was also bad. This argument was based on the proposition that, by making the requirement that they had for returns to be made for the two periods ending 30^th September 1997 and 31^st December 1997 Customs had exhausted their powers to require a return for the earlier period.
38. The argument was said to be based on the provisions of Regulation 25 of the VAT Regulations 1995 (SI 1995 2518). Miss Lonsdale did not, however, succeed in identifying for me the wording in Regulation 25 which supported that contention. At paragraph 46 of the Decision the amendment which had been made to the registration date (from 1^st August 1997 to 1^st August 1996) had been justified. Given a registration date of 1^st August 1996 the effect of Regulation 25 is to require Customs to direct a "first return" in respect of a period beginning with that date: see Regulation 25(1)(b). The fact that they may already (because of a mistake over the date) have made requirements for returns in respect of a later period does not relieve them of that obligation or otherwise disempower them. In any event Regulation 25(1)(c) plainly empowers Customs to vary the earlier requirements both as to period and commencement. Accordingly, I reject that argument.
39. Secondly, Miss Lonsdale argued that the effect of the "amended assessment" was to supersede the second assessment. Accordingly, so the argument ran, once Customs withdrew the amended assessment there was no assessment left in respect of the period.
40. I do not think that this argument can succeed. What Customs conceded at the hearing was that they could not rely on the "amended assessment" because they had no power as such to amend an existing assessment so as to increase the sum assessed. That concession, which appears to me to have been rightly made, leads to the conclusion that the "amended assessment" was a nullity. It did not have some kind of half-life in which it was effective to wash out the second assessment although ineffective to achieve the result at which it aimed.
41. Miss Lonsdale's third argument, which she placed at the forefront of her submissions, was that the civil penalty, insofar as it was based on a notional suppression of takings in this period, was flawed. The argument can be quite simply put. The only VAT in respect of which the appellant was liable in respect of this period was the £6,971.95 assessed by the second assessment. It followed that he could not be said to have evaded payment of any greater sum than this. Yet the quantum of the civil penalty had been based on the premise that the amount of VAT evaded was the £14,284 which had been the subject of the "amended assessment".
42. The Tribunal plainly understood how the figures had been calculated, and were plainly of the view that, as a matter of fact, there had been suppression in the relevant period sufficient to justify the "amended assessment" and therefore the penalty: see paragraphs 60 and 62 of the Decision. The question is whether as a matter of law that is a permissible approach to the penalty. I should add that the point now being taken by Miss Lonsdale was not taken below: see the last sentence of paragraph 62 of the Decision.
43. The relevant statutory provisions are contained in section 60 VATA, which sets out the conditions of liability for, and section 76 which provides the machinery for assessment of, civil penalties. Section 60(1) provides:

"60 VAT evasion: conduct involving dishonesty

(1) In any case where –

(a) for the purpose of evading VAT, a person does any act or omits to take any action, and

(b) his conduct involves dishonesty (whether or not it is such as to give rise to criminal liability),

he shall be liable, subject to subsection (6) below, to a penalty equal to the amount of VAT evaded or, as the case may be, sought to be evaded, by his conduct"

Section 76(1) empowers Customs to assess and notify the penalty. Section 76(3) defines for the purposes of the following paragraphs what is meant by "relevant period", and in relation to a penalty under section 60 it is:

"the prescribed accounting period in respect of which the VAT evaded was due"

Section 76(5) permits, inter alia, a civil penalty assessment to be combined with an assessment under section 73 but the amount of the penalty has to be separately identified. An assessment under section 76 may be made at any time before the expiry of two years from the time "when the amount of VAT due for the prescribed accounting period concerned has been finally determined": see section 77(2).

44. On behalf of Customs Mr Pusey submitted that Customs' ability to assess under section 76 did not depend on its having either first, or at some stage, validly assessed the amount of VAT due in respect of the relevant period. It was true that unless and until such an assessment was made the taxpayer was not liable to pay VAT, but it did not follow from the absence of a valid assessment that the taxpayer had not evaded tax. Here there was no doubt that, on the Tribunal's findings, he had by his failure to make honest returns evaded VAT in the amount purportedly assessed by the "amended assessment".
45. I have not found the point an easy one but have concluded that Mr Pusey's submission is not correct. In my judgment, and contrary to my initial impression, the answer lies in section 77(2). That sub-section deals with the time within which a civil penalty assessment may be raised, and allows such an assessment at any time before the expiry of the 2 years from the time when the VAT due in respect of an accounting period has been "finally determined". The process of final determination will include (but not necessarily be confined to) an assessment. The premise of the sub-section is, therefore, that there will have been an assessment. If a civil penalty assessment can be raised without there having been any determination of the VAT due in accordance with the statutory machinery there would, it would seem, be no time limit applicable to it. I do not think that can have been the Parliamentary intention.
46. Accordingly, in my judgment the civil penalty assessment should be reduced by the sum ascribable to the difference between the amount of the second assessment and the amount of the "amended assessment".
47. Miss Lonsdale's fourth argument (which she described as a fall back position) was that, since the maximum amount of VAT due for the first period was £6,971.95 and the maximum penalty was the same sum, and since the £6,971.95 VAT due had in the event been admitted, the penalty should be subject to the maximum possible mitigation. I would reject that argument. Customs applied a 10% mitigation to the penalty on the basis of the appellant's relative lack of cooperation. That was upheld by the Tribunal. The Tribunal was entitled, in my judgment, not to shut its eyes to the fact that (as it found) the £6,971.95 figure was itself a false one based on a dishonest return.