NCN: [2021] UKFTT 238 (TC)

 

TC08187

 

Appeal number:     TC/2018/04007       

 

VALUE ADDED TAX - whether assessment made to best judgement - yes - whether behaviour deliberate - yes - appeal dismissed  

 

 

FIRST-TIER TRIBUNAL

TAX CHAMBER

 

 

 

	ANNE STEWART	Appellant
	- and -
	THE COMMISSIONERS FOR HER MAJESTY’S	Respondents
	REVENUE & CUSTOMS

 

 

 

TRIBUNAL:	JUDGE ANNE FAIRPO
	MR DAVID MOORE

 

 

 

 

 

Sitting in public at Belfast on 10 March 2020

 

 

Mr Edwards CTA, for the Appellant

 

Mr McKinley, litigator, for the Respondents

 

 

 

DECISION

 

 

Introduction

1.              This is an appeal against a VAT assessment for the periods 11/09 to 11/15 and discovery assessments in respect of income tax and National Insurance Contributions for the tax years 2010/11 to 2014/15, and deliberate behaviour penalty assessments for the same periods.  The assessments, revised to reflect an initial error in the number of days involved, are set out in Appendix A.

Background

2.              The appellant owns and runs a ladies’ shoe shop. The business registered for VAT with effect from 1 September 2009: the appellant stated that she had registered because foreign supplier required her to provide a VAT number before they would supply her. Besides selling shoes and accessories, the appellant also employed a cobbler to provide shoe repairs up until February 2015.

3.              Throughout the relevant period the shop was open Monday-Saturday each week, and is closed on seven days each year (Christmas Day, Boxing Day, Easter Monday and Tuesday, May Day and July 12 and 13).

4.              HMRC visited the premises, unannounced, on 6 November 2015 and took data from the two tills in the shop at that time and spoke to the appellant. From the data retrieved, HMRC concluded that sales had been understated as the cumulative sales total in the till data was higher than the appellant’s declared turnover for the relevant period.

5.              The till data extracted showed cumulative sales totals of £404,895 for one till (the till primarily used for shoes sales) and £323,447 for the other (the till primarily used for shoe repairs). The electronic journal stored in the till, which records all actions performed on the till, covered only the period 3 October 2015 to 6 November 2015. No explanation was given for the short period covered by the journal; the manual for the tills indicated that the till had 3000 lines for electronic journal memory.

6.              As the till data did not provide detailed information that could be used to calculate amounts for each VAT period, HMRC made the assessments on the basis of average takings. HMRC calculated annual average gross takings by taking the two cumulative sales totals, dividing by the number of days since the tills had started to be used and then dividing by 365. It was agreed that an incorrect number had been used for the number of days since the tills had started to be used: the calculation was done on the basis of 2249 days, whereas 2280 days had elapsed between the date on which the tills started to be used (9 August 2009) and the date of the visit (6 November 2015). It was agreed that revised figures would be used if the assessments were upheld. For VAT, this annual figure was used to calculate a quarterly figure which was then applied to each VAT period. The average annual takings calculation was used to produce a net turnover figure which formed the basis of the income tax and National Insurance Contributions discovery assessments.

7.              After some correspondence and a further meeting, HMRC issued assessments for VAT, income tax and National Insurance Contributions, and penalties on the basis of deliberate behaviour. Following a statutory review, the appellant appealed to this Tribunal on 29 May 2018.

Appellant’s evidence and submissions

8.              The appellant contended that the amounts declared to HMRC were correct and that the cumulative sales total data from the tills arose because a former employee, the cobbler, had made repeated “over-ring” errors, confusing the “0” and “00” keys on the till to record (for example) a £700 sale where the actual sale was £7. In addition, the appellant contended that the till supplier had confirmed that the tills were ex-demonstration models and had been used in the showroom prior to purchase by the appellant.

Evidence

9.              The appellant provided a witness statement and gave oral evidence at the hearing.

10.           She explained that she had engaged the cobbler in 2009, to work 16 hours per week. In cross-examination, she said that he worked 20 hours per week. She had engaged a sale assistant, Angela Rice (AR), at the same time to also work 16 hours per week. The appellant worked six days a week in the shop, taking a few hours off when AR was working. AR worked four days a week, and sometimes also worked afternoons. The cobbler had left the business in December 2014 and the appellant ceased to offer shoe repairs at that time. A new sales assistant was engaged in May 2015.

11.           The products sold in the shop are not “high end” shoes, with the most expensive being shoes selling for around £75 and winter boots at around £120. Items are often reduced to clear stock. The appellant aimed for 100% markup but had to discount regularly, up to 60-70%, to turn stock over.

12.           The two tills at the shop were used as follows: one till was used for shoe repairs and miscellaneous items; this till was not frequently used after the business ceased to offer shoe repairs. It was occasionally used by the new sales assistant. The second till was used for sales of shoes, scarves, handbags and gloves. The appellant and (AR) would usually use this till. When mistakes were made entering sales by the appellant or AR, they would try to correct them as customer buying shoes generally required a receipt. However, in cross-examination, the appellant stated that when she made errors these were corrected on the Z reading slip as she did not know how to correct them in the till. She also stated that she did not know what the negative figures in the electronic journal were.

13.           The appellant would run a Z1 report on the till at the end of each day. On the second till, the gross sales listed on the report would generally match the cash and card payments received.

14.           In contrast, the Z1 total sales for the first till would often not be correct. The appellant stated that she very frequently found that the cobbler had made over-rings and he seemed to be confused by the fact that the “0” and “00” buttons on the till were next to each other; he would typically enter a £7 sale as £700. He was not a native English speaker and did not understand what was required. He had a temper and did not respond well to attempts to provide training.

15.           In cross-examination, however, the appellant stated that she had not seen him make such an error as he would generally only use the till when she and AR were busy with other customers. She also said that he might have served customers quite frequently. Customers for shoe repairs did not generally require a receipt and so the over-rings were only apparent at the end of the day.

16.           The appellant and AR would count the tickets from the repairs collected that day in order to calculate the takings for this till; the appellant’s evidence was that this had been AR’s idea. The appellant would score out the Z1 reading total and write in the recalculated amount. The over-rings were not reversed on the till as the appellant did not know that they had arisen until the Z readings were taken.

17.           The daily totals for the tills were recorded by the appellant in a notebook. At the end of the week she would provide her bookkeeper with the figures for the two tills, as adjusted for over-rings. The bookkeeper would provide the necessary information to the appellant’s accountant to produce the VAT returns.

18.           The appellant did not keep the Z1 readings taken daily or the Z2 readings taken weekly. She also did not keep the notebooks in which she recorded the daily totals.

19.           In her witness statement, the appellant stated that the weekly takings on the first till rarely exceeded £500, although in oral evidence she stated that the takings in this till were usually between £500 and £600 per week, although it would be lower in summer and when the cobbler was on holiday.

20.           The tills had been purchased from a local supplier, who had also provided training in their use. The supplier had said that it was possible that the tills had been used as demonstration models in the showroom.The appellant assumed that the supplier had reset the tills when she purchased them but she did not know if this was a full reset.

21.           The supplier would update the till for new brands as the brand and price were recorded on sales. The appellant did not have any detailed knowledge of how the tills worked; she believed that taking the Z reading cancelled the information in the tills and reset the data, and did not know how negative figures in the electronic journal would arise.

22.           When asked why she had not made any comment on the differences between the till data and the declared takings when HMRC had visited, the appellant’s evidence was that she had no reason to believe that the till data retrieved by HMRC would be any different to the amounts declared for tax purposes, as she had not known about the cobbler’s over-rings.

23.           AR also provided a witness statement and gave oral evidence at the hearing.

24.           She stated that she considered the the till readings taken by HMRC would not be accurate, especially the shoe repair till. She herself had made mistakes on the till when she started working at the shop. The cobbler always made mistakes because he seemed to be confused by the “0” and “00” buttons being next to each other on the till.

25.           Her evidence in the hearing was that the cobbler would only serve customers when AR and the appellant were busy with customers, which could be between once a month and twice a week. When told by AR not to use the “00” button, the cobbler replied that he didn’t understand. The cobbler did not use the other, shoe sales, till.

26.           In the hearing, AR said that she had not seen the cobbler make such errors as he only used the till when she was busy with a customer. The hours in which she worked at the same time as the cobbler varied.

Submissions

27.           The appellant did not dispute the methodology used by HMRC as not being to best judgement but, instead, submits that the assumptions made by HMRC as a result of their analysis of the till data are rebuttable and that there is no other firm evidence that sales were supressed. It was submitted that HMRC should have approached the till data with more caution and given greater weight to the appellant’s explanations of the over-rings and the fact that the tills were ex-demo.

28.           The appellant also submitted that HMRC’s assessments produced gross profit figures which were in excess of the average gross margin figure published by a trade industry body, whereas the appellant’s declared figures were closer to that average.

29.           The appellant submitted that the approach to be taken was that in the case of Golden Cube ([2018] UKFTT 488 (TC)), where the judge noted that where there was no evidence of deliberate manipulation of records, the only other options were that there had been mistakes in recording sales such that sales were inadvertently omitted, or there were entries on the till which were not reflective of sales. In this case, there was no indication that sales had been omitted and so the appellant submitted that the only rational explanation was the third: that the tills could not be relied upon because of mistakes in keying sales and/or the use of the tills as demonstration models by the supplier.

30.           With regard to the over-rings, it was submitted that the difference between the till readings and the declared takings could be explained by just three errors wherein a £7 sale was recorded as being £700.

31.           It was submitted that it was thought that the only the Z1 till information had been cleared on the purchase of the tills and that the Z2 information (which included the accumulated Z1 information to date) had not been cleared. It was also submitted that there had to be doubt as to whether HMRC had correctly retrieved information from the tills: the differing day counts, given the appellant’s evidence that she carried out daily Z1 readings, indicated that an error could have occurred. The officer who had carried out the work had been unfamiliar with the till; he since retired and had not given evidence at the hearing although the assessment was based entirely on his work.

32.           The appellant submitted that HMRC had failed to discharge the burden of proof upon them to demonstrate that any alleged errors on returns were the result of deliberate action by the appellant. As such, it was submitted that a significant proportion of the assessments are submitted to be time barred and the penalty assessment is excessive.

33.           It was submitted that HMRC had deemed the appellant to have behaved deliberately because no satisfactory evidence had been provided otherwise: it was submitted that this effectively attempted to reverse the burden of proof. In this case, the burden of proof is on HMRC to show, on the balance of probabilities, that there had been deliberate behaviour on the part of the appellant which had led to an under-declaration of sales. HMRC had deemed the behaviour to be deliberate without actually saying what the deliberate behaviour was. It was submitted that deeming behaviour to be deliberate because no satisfactory evidence has been supplied was a reversal fo the burden of proof. It was contended that the case of Anthony Clynes ([2016] UKFTT 369 (TC)) had held that it was for HRMC to prove, on the balance of probabilities, that the appellant knew that the information provided to HMRC was inaccurate.

34.           The appellant also made submissions as to the validity of an assessment which was dated as the appellant’s final VAT period (1 March 2016 to 11 May 2016), as the appellant had deregistered following the HMRC visit. It was submitted that, as the till information was limited to the periods to the HMRC visit on 6 November 2015, there were no grounds for any assessment to be made by HMRC in respect of this period.

HMRC’s evidence and submissions

Evidence

Officer Quinn and Officer Brine - till systems

35.           For HMRC, Officer Quinn gave oral evidence as a systems compliance officer with experience in dealing with till systems. Officer Brine, who had carried out the till checks and provided a witness statement, had since retired from HMRC and was not available to give evidence. Officer Quinn had adopted Officer Brine’s witness statement but confirmed that he had no direct knowledge of the events described.

36.           Officer Quinn stated that he considered that it would be standard procedure on the sale of a till for all reports to be reset to zero, otherwise any earlier entries would skew the end customer reports. He agreed that, if the clerk readings had not been cleared, they would include anything that had been entered during the period in which a machine was used as a demonstrator. He confirmed that taking a Z1 reading will reset that report but would not clear any other reports on the till. He also confirmed that a Z reading number “1” indicates that it is the first Z reading produced on the machine. Officer Brine’s witness statement also included general information about the reports, noting that a Z2 reading will include the data from all Z1 readings taken since the previous Z2 report had been run. Z2 data will be reset when the Z2 report is run. X reports can also be run which will show the current Z1 data without resetting that Z1 data - they can be used to check information during the day without affecting the end of day report.

37.           Officer Brine’s witness statement confirmed that the make and model of the appellant’s till was not known before the visit and he had no prior knowledge of the tills used by the appellant’s business. He had therefore obtained back office support during the visit to enable him to take the required readings. This was standard procedure. He had produced the reports and completed a questionnaire with the appellant confirming details. The questionnaire noted that: cash expenses were removed from the tills, to pay the window cleaner weekly; approximately ten refunds per month were made through the till; Z reports were run daily around 5pm; daily takings were kept in a notebook.

38.           The information from the till reports included the Z counts, which indicate the number of times that a Z reading had been taken. For one till, this was 1784; for the other, it was 1651. Officer Brine carried out calculations which indicated that the tills had been in use since the appellant started business on 9 August 2009. The clerk report for each till was numbered “1”, and Officer Brine considered that it was reasonable to suppose that these were the first reports of their kind taken from the tills.

39.           Officer Brine also printed the electronic journal from at least one of the tills. This covered the period 3 October 2015 to 6 November 2015. This showed all of the actions on the tills and included, inter alia, a number of negative figures some of which were apparently marked as refunds.

Officer McAuley - VAT assessments

40.           Officer McAuley provided a witness statement and gave oral evidence. She stated that she had visited the appellant’s shop on 6 November 2015 with Officer Brine and another colleague. During the visit, the appellant had run till readings from both tills which matched with the cash and credit card receipts on the day. The appellant had explained that she took a Z reading from each till at the end of the day, and that sales were noted in a hardback sales book kept at the counter. The appellant had also stated that the business had no cash expenses and that drawings were noted elsewhere. VAT returns had been completed by a bookkeeper who had recently resigned; future VAT returns were to be completed by another bookkeeper.

41.           Following the visit, a meeting was held with the appellant and her accountant on 22 April 2016 at the accountant’s office. Officer McAuley stated that the appellant had explained that Z readings and till rolls were given to the bookkeeper to complete the VAT returns but that these were not subsequently kept as the appellant was unaware that she needed to complete return. The appellant provided no explanation as to why the till reports taken at the visit showed different sales amounts to those in her VAT returns. SAGE records and bank statements were provided by the appellant but none of those could confirm the sales amounts declared on the returns as cash takings were not deposited in the bank. The appellant did not provide the journal used to record daily takings.

42.           At the meeting, the appellant stated that the two employees were paid weekly with cash from the takings or, if the takings cash was insufficient, with cash withdrawn from the bank. The appellant had confirmed that approximately 10-15% of sales were paid for in cash. The accounts showed two drawings in cash prior to June 2015, and then weekly cash drawings thereafter. Suppliers were paid by cheque or credit card if the bank account had insufficient funds: stock was purchased twice a year.

43.           Officer McAuley’s evidence was that she had concluded that the under-declaration was deliberate because of the discrepancy between the till reports and the declared takings, and the fact that there was no support for the declared takings as cash takings were not banked, and the till reports showed significant sales of shoes on both tills. The decision was based on the balance of probabilities. Officer McAuley had considered the appellant’s information as to average gross profit margins in the industry but had concluded that this was not useful as she considered that the shoes sold were at the higher end of the industry.

44.           Officer McAuley noted that a VAT penalty had been raised and authorised in time but had not yet been processed.

45.           With regard to the appellant’s submissions as to the final period assessment, Officer McAuley’s evidence was that the final period assessment was not for the dates shown but was, instead, a work-around to deal with the fact that HMRC systems do not permit entries to be made against older individual periods. Accordingly, the assessments for the periods 11/09 to 02/13 were added together and set against the final return. This was, as with the other periods, based on average figures: it was not a bulk assessment, which she stated would have been shown as made for the period “00/00”. This work around meant that the interest charged would be too low, but no manual adjustments had been made to correct that.

Officer Foote - direct tax

46.           Officer Foote provided a witness statement and gave oral evidence. She stated that she had based the income tax assessments on the figures produced from the till analysis, reduced to account for the VAT included in the gross sales figures.

47.           The inaccuracy was considered to be deliberate because the appellant was responsible for accurately recording her sales figures and evidence indicated that this was not done, that the appellant had deliberately failed to keep prime records, and she would therefore have known that the figures provided to her bookkeeper and accountant were not accurate. Officer Foote considered that the difference of over £124,000 would indicate a substantial number of over-rings compared to the number of staff and hours worked. She did not consider it credible that such a level of over-rings would continue for such a length of time.

Submissions

48.           HMRC contended that the appellant had put forward a number of explanations for the discrepancies but had provided no evidence for any of the explanations; for example, neither the appellant nor AR had seen the cobbler over-ring a transaction. In addition, the electronic journal which had been retrieved from the till included negative figures which, it was submitted, showed that over-rings and other errors were cancelled through the till and not left to be corrected manually. The appellant’s contention that the shoe repair till could not have average takings of £1131 per week, from HMRC’s calculations, as the repairs did not exceed £500 per will was not supported as the information from the till showed that it had been used for shoe sales as well.

49.           HMRC also submitted that it was not credible that no action would have been taken to resolve the issue given that it would have taken time to manually correct over-rings by counting shoe repair dockets.

50.           The explanation that the tills had been used as demonstration machines and this was a reason for the higher readings was not, it was submitted, credible. HMRC submitted that a till supply specialist would not have failed to reset the tills completely on sale. It was also noted that the till supplier had not given evidence in the hearing, and that no weight should be given to the “to whom it may concern” letter which had been provided, as this could not be cross-examined.

51.           HMRC submitted that they had made their assessment to best judgement based on the material available to them, being the till information, as the till information and the VAT returns submitted by the appellant could not be reconciled. They had considered all of the evidence available. The appellant had not kept any evidence of sales contrary to the requirements of a12B(3) of the Taxes Management Act 1970, as she had not kept any Z readings or any other contemporaneous evidence of sales made. She had not provided the notebook in which she recorded sales information.

52.           HMRC stated that the trade averages referred to by the appellant were averages across the whole sector and, for example, the effect of providing shoe repairs could affect the position.

53.           HMRC submitted that the decision in Golden Cube referred to by the appellant was not binding on this Tribunal, as it was a First tier decision. That decision also related to a three-day invigilation concerned with the question of the proportion of sales which were standard- or zero-rated supplies. There was no dispute as to the total sales made by the taxable person in Golden Cube.

54.           With regard to the assessment was dated as the final VAT period, HMRC submitted that this was not a bulk assessment but a system work-around to deal with the fact that the VAT system would not permit entries for the earlier periods. The explanation which accompanied the assessment made it clear that the assessments were for the periods 11/09 to 11/15 and did not include an assessment that was for the final period. It was submitted that there was no specified form for a notice of assessment.

Income tax

55.           HMRC submitted that the appellant knew or must have known that the information provided was inaccurate as she had provided no evidence to explain the discrepancies. The penalty was therefore raised on the basis of deliberate behaviour. The penalty had been reduced for cooperation. HMRC further submitted that there were no special circumstances that merited a reduction in the penalty in this case.

Relevant law

56.           s73 Value Added Tax Act 1994 (VATA 1994) provides that HMRC may make an assessment for VAT where a person has failed to keep documents and afford the facilities necessary to verify their VAT returns.

57.           s77 VATA 1994 sets out the time limits which apply in making such assessments: the general rule is that an assessment shall not be made more than four years after the end of the relevant period. However, where a loss of VAT has been brought about deliberately by the taxpayer, an assessment can be made at any time not more than 20 years after the end of the relevant period.

58.           In the case of income tax, s34 Taxes Management Act 1970 (TMA 1970) provides that assessments may be made at any time not more than four years after the end of the year of assessment to which it relates. s36 TMA 1970 extends this time limit in cases of careless behaviour so that assessments may be made not more than six years after the end of the year of assessment to which it relates and, in the case of deliberate behaviour, not more than 20 years after the end of the year of assessment to which it relates.

59.           With regard to the penalties, paragraph 1 of Schedule 24 Finance Act 2007 (FA 2007) provides that a penalty is payable by a person who gives HMRC a document that contains an inaccuracy which leads to an understatement of liability to tax if that inaccuracy was careless of deliberate on that person’s part.

Discussion

60.           The questions before this tribunal were, therefore, whether or not the HMRC assessments for VAT had been made to best judgement (and, consequently, whether the income tax assessments were appropriately made) and, if not, whether the appellant had shown that an alternative figure should be considered to be accurate for VAT purposes.

Whether assessment to best judgement

61.           The appellant stated that they made no submission that the assessments were not made to best judgement. The appellant’s contention was, instead, that there was no inaccuracy in the returns provided to HMRC such that no assessment was appropriate.

62.           The appellant submitted, in summary, that the sales information from the till data was inaccurate and that the information declared to HMRC should be regarded as accurate because:

(1)          there had been frequent over-rings by an employee who did not respond to training; 

(2)          the appellant did not know how to reverse over-rings in the till;

(3)          the till supplier had not reset the tills completely when they were sold to the appellant.

63.           It was submitted that the actual sales made were those reported to HMRC, which had been established by counting shoe repair dockets if the cash and card amounts did not match the Z1 reading.

64.           We do not find the appellant’s submissions credible for the following reasons:

(1)          neither the appellant nor her assistant, AR, had seen the employee make over-rings;

(2)          we do not consider that an employer would retain for several years an employee who made errors in the region of several hundred pounds per week, necessitating a recount of repair dockets each time. We also note that the appellant’s evidence was that this till was also used for sales of dye and accessories such that a recount of repair dockets alone would not necessarily provide an accurate figure for sales;

(3)          the appellant stated in the hearing that she had not made any comment on the difference between the till data and her records at the time of HMRC’s visit because she had not known about the over-rings. This is completely contradictory to her evidence that she corrected these over-rings manually when recording takings each day;

(4)          the information from the electronic journal showed that sales were reversed on the till such that we find that the appellant did know how to reverse incorrect till entries, including over-rings. The appellant’s evidence was also that when there were errors made in shoe sales that these would be reversed in order to provide an accurate receipt to the purchaser;

(5)          we do not consider that a specialist till supplier would not reset a till that had been used as a demonstrator on sale. Further, the “to whom it made concern” letter provided by the supplier indicated that only the Z1 report had been reset. The appellant’s evidence was that she ran a Z2 report once a week, but she made no mention of obtaining an unusually high reading when she ran a Z2 report for the first time. Finally, the Z counts from both tills indicate that the till data had been reset around the time that the appellant acquired them, or shortly thereafter;

(6)          the appellant provided no substantive evidence that the information supplied to HMRC was correct: she kept no Z readings from the tills; she did not bank cash takings; she did not provide any of the notebooks used to record sales information to HMRC, including the notebook in use at the time of HMRC’s visit.

65.           With regard to the discussion as to average gross profit margin figures and whether these supported the appellant’s declared takings or, alternatively, indicated that HMRC’s assessments were too high we note that the trade industry figures provided constituted a single figure per year, being an average across the whole retail sector for each year. There was no range information nor any indication where in relation to the average the appellant’s shop would fall. As such, we do not consider that the information would be of any assistance in determining whether any figures discussed were reliable.

66.           It was submitted for the appellant that Officer Brine had failed to print out the till electronic journal, as required by their internal instructions, which would have provided detailed information which could have been reviewed and that the appellant should not be prejudiced by HMRC’s failure to take the journal readings. It is clear from the evidence exhibited to Officer Brine’s witness statement that he did print out the journal, which covered the period from 3 October 2015 to the date of visit, 6 November 2015 (32 days of data). The date range suggested that the journal had been cleared for some reason as the appellant submitted that the journal memory could hold 3000 Z readings (the copy manual for the tills provided by HMRC indicated the same).

67.           There is a substantial discrepancy between the sales figures produced from the till data and the amounts declared to HMRC. In the absence of any credible explanation from the appellant, we find that, on the balance of probabilities, there were inaccuracies in the information provided to HMRC.

68.           Although the appellant stated that they made no submission that the assessments were not made to best judgement, we have considered the submissions made, taking into account the tests set out in Van Boekel [1981] STC 290, and find no reason to disturb the assessments made by HMRC for VAT purposes. It follows that we also find no reason to disturb the assessments made in respect of income tax and National Insurance Contributions as these were based on the same figures.

Whether the behaviour was deliberate

69.           The appellant submitted that HMRC have deemed the behaviour to be deliberate because no satisfactory evidence had been provided by the appellant, and that this effectively reversed the burden of proof.

70.           The penalty regime in Schedule 24 was introduced by the Finance Act 2007 to provide consistency in the penalty system across a range of taxes, and distinguishes primarily between ‘careless’ behaviour and ‘deliberate’ behaviour. The wording was examined in the case of Anthony Clynes, referred to by the appellant, where the Tribunal concluded that “for there to be a deliberate inaccuracy on a person’s part, the person must to some extent have acted consciously, with full intention or set purpose or in a considered way” (§82).

71.           We note that the Tribunal in that case also considered that “the term “deliberate inaccuracy on a person’s part” can extend beyond this … depending on the precise circumstances, an inaccuracy may also be held to be deliberate where it is found that the person consciously or intentionally chose not to find out the correct position, in particular, where the circumstances are such that the person knew that he should do so … We view the case where a person makes such a conscious choice not to take such steps with the result that an inaccuracy occurs, as no less of a “deliberate inaccuracy” on that person’s part than making the inaccuracy with full knowledge of the inaccuracy.”

72.           In this case, there is a substantial discrepancy between the information stored on the appellant’s tills and the information provided by the appellant to HMRC. The appellant has provided a number of possible explanations but has provided no support for any of those explanations, as set out above. As cash takings were not banked, the appellant’s bank statements do not provide any corroboration for either the declared sales or any other figure for sales.

73.           Some of the appellant’s evidence was inconsistent: for example, in her witness statement she said that the takings on the shoe repair till rarely exceeded £500 but in oral evidence said that the takings on that till would usually be between £500 and £600.

74.           Notably, the appellant has not provided any of the notebooks in which she recorded sales information, and more specifically has not provided the notebook which she was using to record sales information at the time of HMRC’s visit. No reason has been provided to explain why this notebook was not retained and provided to HMRC, given that the appellant would have been aware that HMRC were aware of a discrepancy between the till readings and her declared takings. All that has been provided are Z readings and a page from a notebook with the takings from the tills for a week some time after the HMRC visit. We do not consider that this later information can substitute for the notebook being used at the time of the visit (or earlier notebooks) in any way and, as such, does not demonstrate that the appellant provided accurate information to her bookkeeper before the HMRC visit.

75.           The appellant’s contention that it would not be logical to record sales in the tills but then fail to declare them to HMRC needs to be balanced against her statement that she believed that the Z1 reading reset the information in the tills. Her similar contention that she was trading below the VAT threshold such that it was not logical to be registered and not comply with VAT requirements needs also be balanced against her evidence that she was VAT registered because overseas suppliers would not supply her with goods without a VAT number.

76.           On the balance of probabilities, considering the matters summarised above, we consider that the appellant knew that the figures that she was providing to enable her VAT return to be completed were not accurate, and that she therefore knew that the figures declared to HMRC were not accurate. We find that the behaviour that led to the accuracy was, therefore, deliberate.

77.           We note the appellant’s submissions with regard to the Anthony Cooper decision but consider that, as in that case, our conclusion that there are inaccuracies in the information provided to HMRC means that it is difficult to escape the conclusion that the appellant was aware that there were inaccuracies in the information provided to HMRC.

78.           We also note the appellant’s submissions regarding the Golden Cube decision, but do not consider that the case is of any assistance here. That decision addressed a completely different scenario: the dispute in that case related to the ratio of standard-rated to zero-rated sales and the dispute was as to the invigilation exercise carried out by HMRC which the court considered unrepresentative. There was no dispute in the Golden Cube case as to the amount of the total sales, nor had HMRC alleged that any errors arose as a result of deliberate behaviour.

79.           We find, therefore, that the inaccuracies in the returns were deliberate within the meaning of Schedule 24 of Finance Act 2007. As such, we find that all of the assessments were made in time.

80.           We note that HMRC considered whether or not any special circumstances applied that could reduce the penalty and concluded that there were none. Having considered the matter, we see no reason to disturb that conclusion and also note that the appellant made no submissions on the point.

81.           Decision

82.           In summary, therefore, we find that the VAT and income tax discovery assessments were raised to best judgement and within the applicable time limits. We find that the penalty assessment was correctly raised.

83.           The appeal is therefore dismissed.

84.           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.

 

 

ANNE FAIRPO

 

TRIBUNAL JUDGE

 

RELEASE DATE: 28 June 2021

 

APPENDIX A - ASSESSMENTS APPEALED

 

VAT assessments:

 

Discovery assessments and penalty assessments (income tax):

Period	Date issued	Legislation	Initial assessment	Revised assessment
2010/11	25 April 2017	S29 Taxes Management  1970	£5,691.88	£5,309.40
2011/12	25 April 2017	S29 Taxes Management Act 1970	£4,596.67	£4,208.36
2013/14	25 April 2017	S29 Taxes Management Act 1970	£227.12	£0
2014/15	25 April 2017	S29 Taxes Management Act 1970	£4,449.24	£3,073.16
2010/11,2011/12,2013/14 & 2014/15	25 April 2017	Schedule 24 Finance Act 2007	£6,325.90	£5,382.61