Martin Peter Byrne and Ellena Byrne t/a EVA v The Commissioners for HMRC

An oral decision dismissing these appeals was given at the public video hearing on 16 May 2023, followed by a short decision notice. The Appellants subsequently requested full written findings and reasons, which are now given.

The Appellants were registered for VAT from 1 December 2011 as a partnership trading as “Eva”. The business activity was “retail ladies fashions”.

The Appellants appeal against:

Penalties were also imposed, but the Appellants withdrew their appeal against the penalties by notice dated 19 September 2020.

It was confirmed to the Tribunal by both parties at the outset of the hearing as follows. The outcome of the direct tax appeals (paragraph 3(2) and (3) above) will be determined automatically by the outcome of the VAT appeal (paragraph 3(1) above). As there is nothing further in the direct tax appeals that needs independent consideration by the Tribunal, only the VAT appeal needs to be considered by the Tribunal.

At the hearing, the Appellants’ representative, Mr Chong, furthermore accepted that the Appellants had underdeclared their cash sales in the VAT periods in question, and said that the vast majority of what HMRC says in this appeal is undisputed. The Appellants’ issue is with the quantum of the VAT assessment. Mr Chong said that the Appellants should expect to pay a substantial sum, but disputed that it should be anywhere near the figure assessed by HMRC.

As will be further explained below, the sole issue presented to the Tribunal at the hearing concerns the percentage of the business’s sales that were cash sales, as opposed to card sales.

HMRC selected the Appellants’ business for a VAT enquiry as merchant acquirer data available to HMRC showed that its card takings were 111% of the turnover declared in its VAT returns.

HMRC Officer Bebbington carried out a cash test purchase at the business on 25 May 2016, by purchasing a scarf for £15. There was no till, and she was given a handwritten receipt from a small Challenge duplicate book.

Officer Bebbington then carried out an unannounced visit to the business premises on 16 June 2018, where she spoke with Mrs Byrne. Officer Bebbington uplifted the current Challenge receipt book on the premises at the time of the visit. Mrs Byrne said that other Challenge books were not available as they were at home.

The uplifted Challenge receipt book when examined by HMRC was seen to span the period from Monday 13 June 2016 to Thursday 16 June 2016. The sales recorded in it are set out in paragraph 26 below.

From the information contained in this receipt book, Officer Bebbington calculated that 28% by value of the sales recorded in it were cash sales.

On 22 June 2016, HMRC requested Mr Byrne to arrange an inspection of the business’s VAT records, and in particular, of further Challenge receipt books.

On 24 June 2016, HMRC instructed Mr Byrne to retain all Challenge receipt books going forward as it is a legal requirement to keep a record of every individual sale.

On 27 June 2016, HMRC carried out a further cash test sale, by purchasing an item for £15 from the business, and a receipt from a Challenge receipt book was given.

At a meeting with HMRC on 20 July 2016, Mr Chong, in his capacity as the Appellants’ tax adviser in relation to the VAT enquiry, and Mr Byrne, accepted that the Appellants had not declared cash takings in the VAT returns. Mr Byrne said that they only had seven Challenge books covering the period 27 May 2016 to 16 July 2016, the rest having been destroyed in a flood.

HMRC’s analysis of these Challenge notebooks revealed that they did not include the test purchase on 27 June 2016.

On 29 September 2016, HMRC carried out a further cash test sale, by purchasing an item for £15 from the business, and a receipt from a Challenge receipt book was given.

On 14 December 2016, after the business’s VAT returns for 06/16 and 09/16 had been filed, HMRC uplifted the records and again analysed the Challenge books produced. These did not include the cash test sale on 29 September 2016. The cash figures amounted to 1.2% of sales overall.

Following further exchanges between the parties, HMRC issued the VAT assessment under appeal. The assessment determined the amount of undeclared cash takings by assuming that throughout the entire period to which the assessment related (03/12 to 09/17), cash takings were 28% of overall sales, as per the Challenge receipt book uplifted on 16 June 2018 (see paragraphs 10-12 above). The assessment was upheld in an HMRC review conclusion letter dated 15 May 2018.

On 27 June 2018, the Appellant commenced the present Tribunal proceedings, to appeal against the assessment.

On 23 October 2018, HMRC revised the assessment downwards, as a result of now being satisfied that cash takings were 22.4% of overall sales, rather than 28%. This was because further information showed that one of the payments in the Challenge receipt book uplifted on 16 June 2018 that had previously been treated as a cash payment was in fact a card payment.

Mr Martin Byrne says in his witness statement amongst other matters as follows.

A witness statement dated 3 August 2021 of Mr Adrian Hughes, a partner in Batten Hughes Chartered Accountants, states as follows.

A letter from Mr Adrian Hughes to Mr Byrne dated 14 July 2021 states as follows.

The Challenge receipt book uplifted on 16 June 2018 (paragraphs 10-12 above) contains the following sales.

In this table, the cash sales are some 28% of the total sales. As has been noted (paragraph 22 above), this percentage was revised downwards to 22.4% after HMRC were satisfied that one of the cash sales (number 9 in the table) was in fact a card sale.

The period of sales covered by this Challenge book is referred to below for convenience as a 4-day period. The Tribunal takes into account, as the Appellants insist, that it may not be four full days. Mr Chong referred to it as a three and a half day period.

Section 73(1) VATA provides:

In an appeal against a best of judgment assessment under 73(1) VATA, it is open to the Tribunal, in the exercise of a quasi-supervisory jurisdiction, to find that the assessment was not raised to the best of HMRC’s judgment and should never have been made at all, for instance on the ground that it was reached dishonestly or vindictively or capriciously, or was a spurious estimate or guess in which all elements of judgment were missing, or was wholly unreasonable (see Mithras (Wine Bars) v HMRC [2010] UKUT 115 (TCC) (“Mithras”) at [9]-[11]).

In making a best of judgment assessment, HMRC are required to consider fairly all material put before them by the taxpayer, but are not required to make investigations so long as there is some material on which they could reasonably base an assessment. Nevertheless, if HMRC do make any investigations, they must take into account the material disclosed by that investigation. (Van Boeckel v Customs and Excise Commissioners [1981] STC 290, 292g-293a.)

In cases where HMRC were entitled to make the assessment, and the challenge is to the quantum of the assessment, the Tribunal has a full appellate jurisdiction. It can consider any further evidence or arguments presented by the parties in the appeal and reduce the amount of the assessment, thereby substituting its own view on quantum for that of HMRC. (See Mithras at [7], [16]-[21]).

In an appeal against quantum, generally the burden lies on the taxpayer to establish the correct amount of tax due. The HMRC assessment is prima facie right, and remains right until the taxpayer shows not only that it is wrong, but also shows positively what corrections should be made in order to make the assessment right or more nearly right. There is an element of guess-work and an almost unavoidable inaccuracy in a properly made best of judgment assessment. It is therefore not enough for an appellant to show shortcomings in the HMRC methodology. The Appellant must by evidence and argument positively show a methodology and/or figures that are more reliable than those used by HMRC in making the assessment, even if the Appellant’s methodology and figures may also have their own inevitable shortcomings. (See Khan v Revenue and Customs [2006] EWCA Civ 89 at [69].)

The Appellants rely on the following provisions of HMRC internal manuals:

The figures used by HMRC in this case are neither fair nor representative.

The quantum of the assessment is not credible.

Amendments can be made to the methodology applied by HMRC to achieve a fairer and more reliable figure for the assessment.

While a 4-day period may not be an ideal review period, there is no more representative period that could have been used by HMRC as the review period for making the assessment.

If the HMRC assessment was too high, the Appellants could have produced the missing books, so that the assessment could have been based on a full and accurate set of records for a longer test period. The Appellants were given ample opportunity to produce accurate figures going forward. HMRC question why they have not done so.

Without seeing a full and accurate set of records, it is not possible to say whether items 31 and 32 were unusual sales.

The assessment was based on the whole of the figures for the available review period. It is not disputed by the Appellant that all of the sales in the table at paragraph 26 above occurred, and that 22.4% by value of those sales were cash sales. In the absence of complete records for a longer period, HMRC can only act on the information that they have. The figures from the 4-day review period are what they are, and show what they show. Manipulating them to produce a different result does not make them more reliable.

In directions issued on 10 June 2021, the Tribunal required the Appellants by July 2021 to deliver a list of documents on which they intend to rely in the appeal, and to deliver a skeleton argument not later than 21 days before the hearing. Contrary to these directions, the Appellants have produced no skeleton argument. At the hearing, the Appellants are relying on claimed facts for which no evidence has been provided, and make arguments of which HMRC has not been given prior notice in a skeleton.

HMRC accept that the business was overdrawn, but is unable to accept other facts contended by the Appellant, such as the amount for which the business was sold, or the amount of the overdraft or overdraft interest, or the value of the business. The witness statement of Mr Hughes (paragraph 24 above) does not say what the lease for the business premises was sold for. The letter from Mr Hughes (paragraph 25 above) is not an actual valuation of the Appellants’ business at the relevant time, but merely a general statement that a given level of profits would “suggest” that a business has a certain value.

It is not inherently incredible that the Appellants could have earned the amounts that the HMRC assessments conclude that they earned. Furthermore, businesses can be sold for any reason. The mere fact that a business is sold does not mean that it is not profitable.

The Appellants have not discharged their burden of establishing a more reliable figure for the assessment.

This is not one of those rare cases in which the Tribunal, in the exercise of a quasi-supervisory jurisdiction, can find that the assessment was not raised to the best of HMRC’s judgment and should never have been made at all (see paragraph 30 above). The Tribunal is satisfied that HMRC when making the assessment acted in good faith, and sought to apply a reasonable methodology to determine quantum. The Appellants identify no other review period that could be used as the basis for the assessment (paragraphs 37(1) and 38(4) above). It is not disputed that the figure of 22.4% for cash sales correctly reflects the information from the review period.

That being the case, the HMRC assessment will stand, unless the Appellants discharge the burden of showing positively what corrections should be made to it in order to make it right or more nearly right (paragraphs 32-33 above).

The Appellants rely on the internal HMRC guidance at paragraph 34 above. HMRC internal policy and guidance is not binding on the Tribunal, but will be considered by it where appropriate, in the interests of consistency of treatment of different taxpayers. However, to the extent that the internal HMRC guidance merely reflects the case law at paragraphs 30-33 above, it adds nothing of substance. To the extent that it indicates how HMRC should conduct enquiries and make assessments, it deals with circumstances that are different to those pertaining in a Tribunal appeal against the quantum of an assessment. The Tribunal does not conduct enquiries and make assessments. It makes decisions based solely on the evidence and arguments put before it by the parties. It does not make an initial decision as to what the assessment should be, but rather determines whether an appellant has established a more reliable figure than that contained in an assessment that has already been made.

The Tribunal rejects the Appellants’ argument that the 4-day review period on which the assessment was based was unfair and unrepresentative. Both sides agree that there is no other period that could have been used as the review period (paragraphs 37(1) and 38(4) above). Furthermore, the Appellants have not identified any other methodology for making the assessment. The Appellants’ own proposed methodology takes as the starting point the same review period that was used by HMRC, and merely proposes modifications to it. The challenge based on HMRC’s use of a review period of only 4 days therefore fails.

The Tribunal also rejects the Appellants’ argument that items 31 and 32 stand out as obviously highly unusual cash transactions that must necessarily distort the percentage of cash sales in a way that makes the data unrepresentative unless an adjustment is made to remove this distortion.

The Appellants argue that cash transactions tend to be for smaller amounts, while larger sales tend to be card transactions. That may intuitively appear to make sense. However, there is no evidence before the Tribunal that this is in fact the case. Furthermore, even if it were to be assumed that this is generally true, there would still be some small payments that are made by card, and some large payments that are made by cash. It would be impossible to know, in the absence of evidence, what percentage of large payments are made by card or by cash, and what percentage of small payments are made by card or by cash.

There is no evidence before the Tribunal, for instance, of what percentage of sales of a business of this kind (or even of retail businesses generally) are typically paid by card or cash. There is also no evidence of the percentage of large payments that are typically paid by card or cash, or of the percentage of small payments that are typically paid by card or by cash.

The table at paragraph 26 above shows that more sales were made by card than by cash (both numerically and by value), and that the cash sales were on average for smaller amounts than the card sales. Thus, without any modification at all, the table is consistent with any intuitive assumption that cash transactions tend to be for smaller amounts, while larger sales tend to be card transactions, and that more sales in a fashion shop would be by card than by cash.

However, this does not of itself mean that a business of this kind would never have any larger sales that are paid for by cash, or that two larger cash payments over the course of a 4-day period would be unusual or unrepresentative. Any such conclusions would need to be supported by specific evidence.

The Appellants argue that there is no other instance in the 4-day review period in which there have been two consecutive large cash payments, and that this in itself shows that they are wholly out of the ordinary. The Tribunal does not accept this. If a dice is rolled many times, and the number six appears twice in a row after a long period of no sixes being thrown, this does not prove that something wholly or unusual or exceptional has happened.

The Appellants argue that the review period does not include the busiest days of the week. However, even if it were to be assumed that there are more sales on other days of the week, there is no evidence that on those other days, a smaller percentage of sales are by cash.

The Tribunal accepts that items 31 and 32 have a very significant effect on the percentage figure of sales that are cash sales. However, in the absence of any relevant evidence, the Tribunal is unable to conclude, on the basis of mere intuition, that items 31 and 32 must necessarily have the effect of distorting the accuracy of that figure, or that removing items 31 and 32 would necessarily lead to a more accurate figure.

The figures from the 4-day review period may be of limited reliability. However, the Appellant has produced no evidence to establish the claim that removing items 31 and 32 would make the figures more reliable. An appeal to intuition cannot suffice.

The Tribunal also rejects the Appellants’ argument that the amount of the HMRC assessment is not credible, in the sense that the Appellants could not possibly have been making as much money as they would have been if the HMRC assessment was correct.

At the hearing, Mr Chong gave various figures as to how much the Appellants would have been earning if the assessment was correct, and said that these would put both of the Appellants and their household income in the top few percentage points of earners in the UK.

The Tribunal questioned the basis for figures given by Mr Chong. This led to somewhat lengthy and unproductive exchanges between the parties and the Tribunal in relation to the amount of the Appellants’ earnings.

HMRC accepted that on the HMRC figures, the revised taxable profits of the partnership (after payment of business expenses and VAT) were £87,837 in 2012-13, £99,161 in 2013-14, £88,138 in 2014-15, and £97,549 in 2015-16. The profits of each of the partners would have been half of those amounts.

However, there is no evidence before the Tribunal of kind of profits that might generally be generated by a business of the type owned by the Appellants. In the absence of evidence, the Tribunal cannot conclude that this level of profit by a business of this kind is inherently not credible.

The significance of the level of the Appellants’ earnings indicated by the HMRC assessment is said by the Appellants to be as follows.

First, it is said that it is not credible that the Appellants were such high earners given that they led modest lifestyles (paragraph 23(7) above). It was argued in particular that prior to the commencement of the HMRC enquiry, the Appellants were not paying tax on cash sales and were not paying the VAT on those sales to HMRC, such that the cash in their pockets would have been equivalent to a far higher gross income still. It is said that it is not credible that the Appellants were ever in receipt of such large amounts. The Tribunal does not accept this argument.

Secondly, it is said that if the HMRC assessment was correct, the business would have had a value of some £300,000. It is said that if the business was worth this much, it is not credible that the Appellants would have closed it down, and sold the residual lease for £35,000 to someone who wanted to use the premises for an unrelated purpose, thereby receiving nothing for the goodwill. It is said that it is also not credible that the Appellants would have failed to find a buyer for a business worth £300,000 when it was advertised for sale for only £75,000. The Tribunal does not accept this argument.

The Appellants have therefore not discharged their burden of establishing by evidence a more reliable figure than the figure in the HMRC assessment.

This document contains full findings of fact and reasons for the decision. Any party dissatisfied with this decision has a right to apply for permission to appeal against it pursuant to Rule 39 of the Tribunal Procedure (First-tier Tribunal) (Tax Chamber) Rules 2009. The application must be received by this Tribunal not later than 56 days after this decision is sent to that party. The parties are referred to “Guidance to accompany a Decision from the First-tier Tribunal (Tax Chamber)” which accompanies and forms part of this decision notice.