-
VALUE ADDED TAX — restaurant — appeal allowed in part in principle and parties left to determine correct amount of tax — parties unable to agree — whether period removed from assessment as the result of the decision in principle to be taken into account in recalculating remainder of assessment — no — assessment increased in respect of some periods and reduced or discharged in respect of others
MANCHESTER TRIBUNAL CENTRE
AKBARS BALTI RESTAURANT LIMITED Appellant
- and -
THE COMMISSIONERS FOR
HER MAJESTY'S REVENUE AND CUSTOMS
Respondents
Tribunal: Colin Bishopp (Chairman)
Arthur Brown FCA
Sitting in public in Manchester on 5 September 2007
Richard Barlow, counsel, instructed by Independent VAT Consultants for the Appellant
Nigel Poole, counsel, instructed by the Acting Solicitor for HM Revenue and Customs for the Respondents
© CROWN COPYRIGHT 2007
DECISION
- On 5 September 2006 our decision in principle in this appeal was released. We determined, in summary, that the results of observations carried out by the Respondents' officers at the Appellant's restaurant led to the conclusion that, although there was substantial under-declaration of takings in January and February 2000, the measure of under-declaration detected in July 2000 was so small that it did not warrant the conclusion that the Appellant's return for that period was incorrect. We decided that assessments made by the Commissioners in order to recover the under-declared tax should be discharged, so far as they related to prescribed periods 07/00 and 10/00, but upheld in principle in respect of the remaining periods which had been assessed, 04/99 to 04/00 inclusive. We left the parties to recalculate the tax due in the light of our findings. Unfortunately they were unable to do so and, exactly one year after our decision in principle was released, they attended before us (by their counsel Richard Barlow and Nigel Poole) to make further observations on the course we should adopt in ourselves determining the correct amount of tax.
- In our earlier decision we recorded that we accepted the substance of the evidence of the observations which we had heard, but we considered that the conclusions which should be drawn from that evidence should be rather more favourable to the Appellant than the conclusions which had been drawn by the assessing officer, and we also corrected some small errors of arithmetic, as we perceived them. The outcome was that we determined that, on the first day on which observations were carried out, the Appellant declared only 42.67 per cent of the restaurant's true takings for the day, and on the second day, 26 February 2000, 61.54 per cent. On the third day of observations, however, the difference between the expected takings and those declared was only 6.63 per cent. The expected takings had been calculated, not from meal tickets or other primary evidence, but by determining the number of customers, an exercise, however carefully it is undertaken, which is open to some degree of error, and then multiplying the number so found by the estimated average cost of a meal, also an imprecise measure. We had no doubt that the assessing officer had approached her task carefully, but we concluded that the declared takings were so close to those expected that the Appellant had discharged the burden of satisfying us that, on that day, there was no reliable material on which it could be determined that the takings were not correctly declared—in other words, we were satisfied that the declaration was correct.
- It will be observed that, as we recorded in our earlier decision, the under-declarations follow the unusual pattern of diminishing with the passage of time. We determined that, by July 2000, under-declaration had ceased and we therefore discharged the assessments for periods 07/00 and 10/00. However, the assessments had all been calculated by the same means, of taking the average under-declaration detected on the three days of observations, and applying that average to the declared takings in each of the seven prescribed periods. One course which occurred to us was to adopt the same approach, though using the average of our own, lower, calculations of the under-declarations on the three days. That approach would have had the merit of simplicity, but it did not appear to us to be the right course once we had concluded that the assessments for 07/00 and 10/00 should be discharged; it would be inappropriate and inconsistent with that conclusion to determine nevertheless that some tax was due for those periods. For those reasons we decided that the assessments for the two last periods should be discharged.
- We were conscious, when reaching that conclusion, that the calculation of the under-declared tax for the remaining five periods would need to be reconsidered and that, if the proper course was simply to apply the average of the under-declarations in January and February to the declarations in those periods in order to determine the correct tax, the assessments would be increased, and by a significant amount. It is within our power to make a direction to that effect: see section 84(5) of the Value Added Tax Act 1994; and there is support for the proposition that the tribunal should not shrink from doing so in the observations of Carnwath LJ in Customs and Excise Commissioners v Pegasus Birds Ltd [2004] STC 1509 at [38]. We were reluctant, however, to adopt that course without first giving the parties the opportunity of agreeing the correct amount of tax themselves and if, as has happened, they were unable to do so, without allowing them to address us further.
- Mr Poole's position was the simple one that, the Appellant having secured a reduction in respect of some of the periods assessed as a result of our findings, it could not expect to be sheltered from the inevitable additional but adverse consequences of the same findings. If, as we had found, there was substantial under-declaration in the periods up to 04/00 (in which the January and February observation days fell) the tax due for those periods should be assessed by reference to the degree of under-declaration which had been detected. He therefore asked us to adjust the assessments by applying the average under-declaration in January and February 2000, as we had determined it, to the declarations.
- Mr Barlow's position was altogether more complicated. He did not argue, nor was there any evidence to support the proposition, that the under-declarations had been confined to period 04/00, and he accepted that some adjustment to the amounts assessed in the five remaining periods should be made. However, he said, we should not only include the low rate of under-declaration detected in July, but should follow through our own finding that it fell within a reasonable margin of error, by deducting the rate of declaration calculated for July from the average. We should mention at this point that the (weighted) average of the calculated under-declarations in January and February, expressed as the ratio of under-declared to declared takings (a different but, we think, simpler approach from that adopted in the tables in our earlier decision), is 83.61 per cent, the average if all three days are taken is 44.94 per cent and, if we adopted Mr Barlow's suggestion, it would be 37.84 per cent.
- In our view Mr Barlow's argument goes too far. As we indicated in our earlier decision, there was no evidence or other information before us from which we might conclude why, or precisely when, the Appellant began to declare its takings correctly. Likewise there was nothing before us to indicate that January and February 2000 were atypical and, as we have mentioned, Mr Barlow did not suggest to us that there was any reason why we should not conclude that there had been under-declarations of a similar magnitude in the earlier periods. Indeed, if there was a downward trend, it may be that the scale of under-declaration in the earlier periods was even higher, but we do not think we can properly come to such a conclusion on the strength of the two days of observations with which we are left.
- We have little doubt that the calculated rate of under-declaration in January and February 2000 is wrong—it would be remarkable, in fact, if it were right within five or even ten per cent—but it is as likely that the calculated rate understates the truth as that it overstates it. We do not, therefore, accept Mr Barlow's argument that we should adjust the average, either of those two days or of all three, by deducting the rate found in July. Nor do we think it is appropriate to take the average of the three days. The last day has already been taken into account by our discharging the assessments for the final two periods, upon the basis that by then under-declaration had ceased. A period in which no under-declaration was made can be no guide to the scale of the under-declarations in periods in which they were made. It is, as Pegasus Birds made abundantly clear, for an appellant seeking to challenge an assessment of this kind to advance evidence which supports an assessment of a different amount. This Appellant has not done so and we are left only with evidence that, in January and February, the scale of under-declaration amounted to 83.61 per cent, and no material on which we might conclude that the Commissioners were wrong to base the assessments for that period (04/00) and the preceding periods on the assumption that there had been under-declarations at that rate throughout. As we have said, we do not know why the under-declarations ceased, or precisely when, but the fact that they did casts no light on what occurred between 1 February 1999 and 30 April 2000, the period covered by the remaining assessments.
- We are satisfied, therefore, that the assessments for those periods must be re-determined by reference to the weighted average under-declaration in January and February 2002, as we have recalculated it. We have set out our calculation of the correct amounts in a table annexed to this decision, and we determine the aggregate amount of tax for which the Appellant is properly assessed, in accordance with that table, at £62,278.68, representing overall a reduction of £8,402.32 from the total of £70,681 for which it was originally assessed.
- We give the parties liberty to make an application for a direction in respect of costs.
COLIN BISHOPP
Chairman
Release date: 5 October 2007
MAN/02/0179
| Observations |
Observations |
Observations |
Observations |
| |
19.1.00 |
26.2.00 |
21.7.00 |
| Expected |
£ 2,132.86 |
£ 3,558.95 |
£ 3,392.95 |
| Declared (A) |
£ 910.00 |
£ 2,190.00 |
£ 3,168.00 |
| Difference (B) |
£ 1,222.86 |
£ 1,368.95 |
£ 224.95 |
| B ... A |
134.38% |
62.51% |
7.10% |
| straight average |
|
98.44% |
68.00% |
| weighted average |
|
83.61% |
44.94% |
| |
|
|
|
| Assessments |
Assessments |
Assessments |
Assessments |
Assessments |
| Period |
declared |
assessed |
revised |
change |
| 04/99 |
£ 9,581.51 |
£ 5,353.00 |
£ 8,011.10 |
£ 2,658.10 |
| 07/99 |
£ 16,079.47 |
£ 8,983.00 |
£ 13,444.04 |
£ 4,461.04 |
| 10/99 |
£ 13,903.64 |
£ 7,767.00 |
£ 11,624.83 |
£ 3,857.83 |
| 01/00 |
£ 18,406.43 |
£ 10,283.00 |
£ 15,389.62 |
£ 5,106.62 |
| 04/00 |
£ 16,516.08 |
£ 9,227.00 |
£ 13,809.09 |
£ 4,582.09 |
| 07/00 |
£ 22,631.77 |
£ 12,644.00 |
£ - |
-£12,644.00 |
| 10/00 |
£ 29,396.92 |
£ 16,424.00 |
£ - |
-£16,424.00 |
| totals |
£ 126,515.82 |
£ 70,681.00 |
£ 62,278.68 |
-£ 8,402.32 |
