BAILII [Home] [Databases] [World Law] [Multidatabase Search] [Help] [Feedback]

First-tier Tribunal (Tax)

You are here: BAILII >> Databases >> First-tier Tribunal (Tax) >> Golden Cube Ltd v Revenue & Customs (VAT - ASSESSMENTS : Other) [2018] UKFTT 488 (TC) (20 August 2018)
Cite as: [2018] UKFTT 488 (TC)

[New search] [Contents list] [View without highlighting] [Printable PDF version] [Help]





Appeal number: TC/2016/05440


VALUE ADDED TAX – restaurant – split between standard-rated and zero-rated supplies – whether output tax understated – appeal allowed















- and -





















Sitting in public at North Tyneside Magistrates Court on 12 June 2018


Mr Alan Watson of Watson Associates, Accountants appeared for the Appellant


Mr Gareth Hilton of HM Revenue & Customs appeared for the Respondents









1.       The appellant trades as a franchisee of Subway from shop premises in South Shields. This appeal concerns an assessment to Value Added Tax dated 20 April 2016 in the sum of £47,875 covering periods 05/12 to 01/15 (“the Assessment”). The Assessment was made following an enquiry by HMRC in which the officer conducting the enquiry concluded that the Appellant had been incorrectly zero-rating certain supplies which ought to have been standard-rated.

2.       HMRC also raised a separate assessment dated 16 May 2016 in the sum of £14,042 which comprised under declared output tax on reverse charge supplies made by the appellant for periods 11/14 and 02/15 and disallowed input tax for periods 05/12 to 11/15. There was no appeal against the sum assessed for output tax on reverse charge supplies. In the light of further material provided by the appellant shortly prior to the hearing HMRC agreed that the assessment to input tax should be set aside. In the circumstances this appeal is solely concerned with the Assessment.

3.       In the course of opening, Mr Alan Watson who appeared for the appellant acknowledged that he was not challenging whether the Assessment had been made to best judgement. I bear in mind the guidance given by the Court of Appeal in Customs & Excise Commissioners v Pegasus Birds Ltd [2004] EWCA Civ 1015 where at [38] Carnwarth LJ as he then was stated as follows:

The Tribunal should remember that its primary task is to find the correct amount of tax, so far as possible on the material properly 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 judgment at the time of the assessment.

4.       In the light of Pegasus Birds, the principal issue on the appeal therefore is whether the Assessment is excessive. The burden is on the appellant to satisfy me on the balance of probabilities that the Assessment is excessive. The appellant contends that standard-rated and zero-rated supplies have been properly identified and output tax correctly accounted for. Mr Watson submits that the Assessment should therefore be reduced to nil. Mr Gareth Hilton who appears for the Respondents seeks to maintain the Assessment.

5.       I heard evidence on behalf of the Appellant from Mr Stephen Wood, a director and shareholder in the appellant; Ms Kay Tiffin and Ms Sarah Duke, both employees of the appellant; and from Mr Watson, the appellant’s accountant. On behalf of HMRC I heard evidence from Mr Darren Jordan, the officer of HMRC who conducted the enquiry and made the Assessment. All witnesses provided witness statements and gave oral evidence. On the basis of that evidence and the documentary evidence before me I make the following findings of fact, all by reference to the balance of probabilities.


Findings of Fact

6.       The appellant’s shop is located in a back lane some 50 yards off a main street in South Shields. It is next door to a loading bay operated by Argos and has no view as such. It is open Monday to Saturday, 7am to 8pm and on Sunday, 10am to 6pm. The menu principally offers hot and cold sandwiches known as “subs” and salads. There is only one hot filling, which is meatballs. However any sandwich may be toasted at the customer’s option using a toaster oven. Customers at the shop include lunchtime office workers, shoppers and children.

7.       The appellant has been registered for VAT since 2011. Mr Jordan first visited the appellant on 11 August 2015. He met with Mr Wood and Ms Parry, who was then the shop manager. He explained to them that the proportion of standard-rated to zero-rated sales was lower than he would have expected. It was agreed that Mr Jordan would carry out an invigilation exercise. The invigilation exercise involved two officers attending the shop premises to observe the operation of the till and to record details of individual sales transactions.

8.       Invigilations took place on 12, 15 and 21 October 2015 which were a Monday, a Thursday and a Wednesday respectively. Mr Jordan was present on each date together with another officer. The officers observed all sales on each day during the period of each invigilation, which was 10.20am to 2.50pm on 12 October, 10.30am to 2.50pm on 15 October and 10.35am to 3.05pm on 21 October. The times were chosen by Mr Jordan so as to cover the busy lunchtime trade. On 21 October 2015 the till system stopped working at approximately 12 noon and staff recorded sales manually, categorising sales between eat in/eat out and between hot/cold food.

9.       The shop staff provided copies of individual till receipts to Mr Jordan for all sales during the period of each invigilation. Mr Jordan was satisfied and I find that during the period of each invigilation all sales were correctly entered into the till and for all of those sales the correct VAT liability was applied. In broad terms standard rated sales comprised food consumed on the premises and hot takeaway food, and zero rated sales comprised cold takeaway food. For the period when the till was not working all sales were correctly categorised by the staff.

10.    Mr Jordan analysed the results from the three invigilations as follows:


Total Gross Sales


SR Sales



ZR Sales



Proportion SR Sales


Proportion ZR Sales








12 Oct 2015






15 Oct 2015






21 Oct 2015



















11.    Mr Jordan identified that the proportion of standard rated sales during the invigilation exercise was much higher than the proportion shown by the appellant’s VAT returns for previous periods. The average proportion of standard-rated sales (including VAT) to gross sales over the 3 days of invigilation was 82%. The gross sales and the proportion of standard-rated sales (including VAT) to gross sales for VAT periods 05/12 to 11/15 were as follows:



Gross Sales


SR Sales



SR Sales



Gross Sales


SR Sales



SR Sales











































































12.    When the appellant first registered for VAT its first VAT return for period 08/11 showed 79% standard rated sales, period 11/11 showed 71% and period 02/12 showed 54%. These figures were provided by Mr Jordan and I accept them. He acknowledged however that unlike the figures above they represented the proportion of standard-rated sales (excluding VAT) to total sales net of VAT.

13.    Following the invigilation exercise there was a meeting at which Mr Jordan gave an overview of his findings. Mr Wood expressed concern as to whether the invigilation exercise was representative. In particular, he was concerned that there were only three visits and each visit lasted only until mid-afternoon. Mr Jordan offered to carry out further invigilations but at that stage Mr Watson wanted to review Mr Jordan’s findings. Mr Jordan left details of the invigilation exercise and Mr Watson intended to respond by the middle of January 2016.

14.    In the absence of any response, Mr Jordan wrote to the Appellant on 10 February 2016 enclosing preliminary calculations showing that additional VAT of £47,874 would be due if the average proportion of standard rated sales of 82% shown by the invigilation exercise were applied to the appellant’s turnover for VAT periods 05/12 to 11/15. The appellant was given an opportunity to provide comments or additional information. Mr Jordan also stated that he was happy to undertake further invigilations if the appellant did not consider the results were representative.

15.    There was no response by the appellant to that letter, but it appears a letter from Mr Watson crossed with it in the post. Mr Watson wrote by letter dated 8 February 2016. Mr Watson took issue with a computational aspect of the invigilation figures left by Mr Jordan. Mr Jordan replied on 3 March 2016 in a letter to the appellant, maintaining his method of calculation. Mr Jordan again offered the opportunity for further invigilations.

16.    There was no response to Mr Jordan’s letter. In his evidence Mr Watson stated that he had not accepted the offers of further invigilations because at the time those offers were made the amount of VAT being sought by HMRC had not been quantified.  He said that if the sum sought had been modest, the appellant may have taken a commercial view not to challenge the calculation. In fact, it is clear that by the time of the second offer of further invigilations made on 10 February 2016 the amount of VAT had been quantified. Unfortunately, however there was a breakdown in communication between Mr Wood and Mr Watson. Mr Jordan’s correspondence was being sent to Mr Wood and it was not being copied to Mr Watson or forwarded to him.

17.    On 29 March 2016 Mr Jordan issued a notice of assessment to the appellant. I understand that a computer generated notice of assessment was also sent on 20 April 2016. The Assessment was made pursuant to section 73(1) Value Added Tax Act 1994. A copy was sent to Mr Watson and again it appears to have crossed with a letter from Mr Watson to Mr Jordan dated 23 March 2016 in which Mr Watson effectively argued that as the till was operated correctly during the invigilations, the VAT returns must be accurate. In a response dated 15 April 2016, Mr Jordan maintained that standard-rated sales had not been entered into the till correctly at other times.

18.    On 16 May 2016 Mr Watson wrote suggesting that the invigilation exercise was not representative because it did not cover evening trade or weekends, and it was at a colder time of year when more customers might order hot food. He invited Mr Jordan to carry out two further invigilations. However, at this stage Mr Jordan appears to have considered that as the Assessment had been made, it was not appropriate to carry out further invigilations. In the event the appellant sought a review of the Assessment and the Assessment was confirmed in a review dated 12 September 2016 without any further invigilations taking place.

19.    On this appeal, the appellant challenges the Assessment broadly on the basis that the invigilations are not representative because they were carried out at the same time of day and year and they did not include weekends and evenings. Further, the Assessment did not take into account periods when the appellant’s toasting machine was broken so that hot food could not be sold. The appellant also contends that the invigilations at least show that sales are entered into the till correctly, there is no evidence to suggest that sales have been deliberately entered incorrectly and there is no evidence to support any errors in the operation of the till.

20.    I now turn to consider how the shop, and in particular the till was operated. Mr Wood does not work in the shop on a day to day basis. The appellant employs staff and supervisors to work in the shop, although Mr Wood does call in once or twice a week.  On those visits he will meet with one of the supervisors, pay any outstanding bills and sort out any problems. He does not serve customers whilst he is there.

21.    The witness statements of Ms Duke and Ms Tiffin were identical. They were drafted by Mr Wood and Mr Watson. It would have been better if those witnesses had been invited to make witness statements using their own words. However, their oral evidence in chief and under cross-examination gave them an opportunity to give their evidence in their own words.

22.    I am satisfied that all witnesses gave honest and, subject to one point below, reliable evidence. Subject to that point I accept the evidence of Mr Wood, Ms Tiffin and Ms Duke.

23.    Ms Tiffin has been employed as a supervisor at the shop since 2010. Her role involves managing stock and organising and supervising staff. She also trains junior employees, including till training which takes place during quieter periods. She was not aware of the difference in VAT treatment of difference types of sales. She had never been instructed to ring sales into the till incorrectly and had never instructed any member of staff to do so. She stated that the sales mix between hot and cold food and between eat in and takeaway could differ wildly depending on the day of the week, time of day and season. She was unable to identify any relevant pattern to the sales. She acknowledged that sandwich orders may be entered incorrectly in to the till, but maintained that such errors would be minimal.

24.    Ms Tiffin stated in her witness statement that there were “many times” when the shop equipment was out of order, including the toasting oven so that toasted sandwiches could not be sold. In her oral evidence she clarified that whilst the toasting oven had been out of order,  it was not a frequent occurrence, maybe 4 or 5 times for one or two days before it was repaired. Eventually a new toasting oven was purchased in 2015 and the shop was without a toasting oven for a week before the new toasting oven arrived. Customers were very understanding, and if necessary their sandwiches would be heated in the microwave, although this did not give a good result.

25.    Ms Duke has also been employed as a supervisor at the shop since 2010, in the same role as Ms Tiffin. She too was not aware of the difference in VAT treatment of difference types of sales. She had never been instructed to ring sales into the till incorrectly and had never instructed a member of staff to do so. She also stated that the sales mix could differ wildly depending on the day of the week, time of day and season.

26.    Mr Wood’s evidence in his witness statement was that there were “many times” when the shop’s toasting machine was out of order so that hot sandwiches could not be sold. The evidence of Ms Tiffin and Ms Duke in their witness statements was to the same effect. I do not regard that evidence as reliable and each witness in oral evidence described the toaster oven as breaking down “occasionally” or words to that effect. I accept that the toaster oven did break down on occasion, but the occasions were not very frequent and would not have had a significant effect on the ratio of standard rated sales to total sales.

27.    The till operated by the appellant is programmed by Subway and linked to Subway’s systems electronically. Mr Wood has not made any adjustments to the till or to its software. The evidence included a “Weekly Inventory and Sales Report” for 17 October 2017. These weekly till reports are generated by the till and retained by the appellant. Mr Watson prepares the VAT returns for the appellant and has been engaged to do so since about 2014. He is supplied with the weekly till reports for that purpose. Mr Watson does not perform any checks on the information provided in the weekly till reports.

28.    The till has a touch screen and uses images to prompt the operator to enter details of the customers’ choice of sandwich roll, fillings and extras. It also prompts whether the sandwich is hot or cold and whether the customer wants to eat in or takeaway. The till is programmed to identify the applicable VAT rate to each transaction depending on the information entered. The supervisors employed by the appellant are trained to use the till by Subway. The supervisors themselves train other employees.

29.    The weekly till report is printed out on Tuesday night each week by a supervisor. The figures are submitted to Subway electronically by the till on Wednesday morning each week, although Subway can monitor information stored in the till at any time. Mr Wood sends a hard copy of the weekly till report to Mr Watson. The weekly till report shows the gross sales in each week, calculates the VAT output tax on supplies and shows the net sales. There is a breakdown of the daily sales. The weekly till report also shows royalties and other sums payable to Subway under the franchise agreement. Effectively, a sum of 12.5% of sales net of VAT is payable to Subway.

30.    The weekly till report also shows opening stock, usage and closing stock of each type of ingredient used in the shop.

31.    HMRC do not assert any positive case as to why the standard-rated sales might be understated and it is not necessary for them to do so. In my view and in the circumstances of this case it is helpful to consider in broad terms what reasons there might be for the invigilations to show 82% standard rated sales whereas previous VAT returns show 44% - 65% standard rated sales.

32.    Mr Hilton for HMRC accepted that there may be three reasons for differences in the proportions of standard-rated sales between the invigilation days and the previous VAT returns:

(1)   Deliberate manipulation of the records.

(2)   Mistakes in the recording of standard rated sales, either in the till entries, in the till programming or in the transfer of figures from the till records to the VAT returns.

(3)   The three days of invigilation are not representative of the previous VAT periods.

33.    HMRC do not assert that the records have been deliberately manipulated. Quite rightly, because there is no evidence to support such an assertion. I can go further. On the basis of the evidence before me I am satisfied that there has been no deliberate manipulation of the records.

34.    There is no suggestion that the till was incorrectly programmed and Mr Jordan accepted that the VAT returns were consistent with the weekly till reports. Obviously errors can occur in making till entries, but the nature of the till with a touchscreen using images to prompt entries means that such errors are likely to be minimal, and would not explain the results of the invigilation. It is notable that on the invigilation days no errors were made. The invigilations took place at the busiest time of day when one might expect mistakes to be made. It may be that the till operators were more careful when the invigilation was taking place. However, when they are given an option between hot or cold, and between eat in or takeaway it seems unlikely that there would be frequent and consistent errors one way or the other. The evidence did not suggest that the till had any default setting which might have led to errors giving rise to a disproportionate value of standard-rated sales.

35.    That leaves the possibility that the periods of invigilation were not representative of the appellant’s trade over the previous VAT periods. I am satisfied that the appellant’s trade may be subject to daily variations, variations arising from the time of day and/or seasonal variations. I acknowledge that the appellant has not carried out any exercise to demonstrate what the effect of such variations might be. Having said that, I must consider whether or not I can be satisfied on the basis of the evidence before me that the invigilation results are explicable by reference to such variations.

36.     Mr Jordan suggested in his oral evidence that such variations were unlikely to explain the differing proportions of standard-rated sales. He relied upon his experience of other Subway shops in Newcastle, Darlington and Consett. There was no direct evidence before me as to the trade of other Subway shops. In the circumstances it would not be fair for me to accept Mr Jordan’s evidence as to the proportion of standard-rated sales at Subway shops in other locations at face value.

37.    I note that the proportion of standard-rated sales in period 11/15 was 72% and was somewhat higher than the previous periods. This was the period in which the invigilations took place. Mr Jordan also told me that periods after 11/15 were much lower than 82% but there were no documents to enable that evidence to be tested. I am not satisfied that there is any trend in the proportion of standard rated sales over the period of the Assessment that helps to explain the invigilation results.

38.    Taking into account all the evidence and my findings based on that evidence, on balance I am satisfied that the most likely explanation for the different proportions of standard-rated sales is that the invigilations were not representative of the VAT periods to which they were compared. I cannot say precisely why that should be the case. It may be because the invigilations were all carried out during lunchtime over 3 days at the same time of year and over a short period of time. I do not criticise Mr Jordan for his choice of sample days and for his decision to concentrate on the lunchtime trade. However, with the benefit of hearing all the evidence I must conclude that it was not representative.



39.    For the reasons given above I am satisfied that Assessment is excessive and that there was no understatement of standard-rated sales by the Appellant in the periods assessed. In the circumstances I allow the appeal.

40.           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.





RELEASE DATE: 20 August 2018

BAILII: Copyright Policy | Disclaimers | Privacy Policy | Feedback | Donate to BAILII