[2016] UKFTT 360 (TC)
TC05115
Appeal number: TC/2014/5426
Gaming machine Licence Duty – Betting and Gaming Duties Act 1981 – recovery of Amusement Machine Licence Duty on unlicensed machines- whether number of licences held by Appellant fell short of the number of machines deployed
FIRST-TIER TRIBUNAL
TAX CHAMBER
|
ARROW LEISURE LTD |
Appellant |
|
|
|
|
- and - |
|
|
|
|
|
THE COMMISSIONERS FOR HER MAJESTY’S |
Respondents |
|
REVENUE & CUSTOMS |
|
TRIBUNAL: |
JUDGE CHARLES HELLIER |
|
GILL HUNTER |
Sitting in public at the Royal Courts of Justice, London on 21 and 22 March 2016
Liban Ahmed of Controlled Tax Management for the Appellant
Michael Paulin, instructed by the General Counsel and Solicitor to HM Revenue and Customs, for the Respondents
© CROWN COPYRIGHT 2016
DECISION
1. This appeal relates to an assessment to Amusement Machine Licence Duty (“AMLD”) made on the Appellant in relation to the number of gaming machines which it provided to B&N Regal (Abingdon) Ltd at that company’s premises, the New Coronet Bingo and Social Club between September 2009 and January 2013.
The relevant statutory provisions
2. Section 21 Betting and Gaming Duties Act 1981 provides that no amusement machine shall be provided for play on premises in the UK unless there is a licence in force for it or it is exempt. Sections 22 and 23 impose a charge on amusement machine licences by reference to the period of the licence and the category of the machines (determined according to the price to be paid for a game and the maximum prize). This appeal concerns category B3 and C machines. From August 2011 the rate of duty for a B3 licence was £215 for a single month and £2405 for 12 months with a scale between the two for intermediate periods.
3. Sch 4A of the Act provides that if HMRC conclude that an unlicensed machine is provided for play in the UK, they may serve a default notice requiring the production of a licence, and if a licence is not produced may grant a default licence. Para 4 provides that where a default licence has been granted HMRC may assess the duty which should have been paid on the licence. Para 5 provides that the assessment may be made on a responsible person, which by para 77 includes the owner of the machine.
The assessment
4. The assessment under appeal is for £19,730. It relates to 14 periods (of varying lengths) during which HMRC say that the appellant provided at New Coronet more category B3 machines than those for which it had obtained licences. The table which follows describes the detail:
Table 1
Period Number |
Dates |
No of B3 machines per HMRC |
No of B3 licences per HMRC |
(Deficit) in licences |
1 |
4/9/09 to 26/11/09 |
7 |
2 |
(5) |
2 |
27/11/09 to 6/1/10 |
7 |
4 |
(3) |
3 |
7/1/10/ to 29/3/10 |
7 |
4 |
(3) |
4 |
30/3/10 to 6/4/10 |
7 |
6 |
(1) |
5 |
7/4/10 to 12/4/10 |
7 |
6 |
(1) |
6 |
13/4/10 to 27/4/10 |
7 |
9 |
- |
7 |
28/4/10 to 26/11/10 |
7 |
7 |
- |
8 |
27/11/10 to 29/3/10 |
7 |
7 |
- |
9 |
30/3/11 to 12/4/11 |
7 |
7 |
- |
10 |
13/4/11 to 12/11/11 |
7 |
5 |
(2) |
11 |
27/11/11 to 29/3/12 |
7 |
5 |
(2) |
12 |
30/3/12 to 12/4/12 |
7 |
5 |
(2) |
13 |
13/4/12 to 26/11/12 |
7 |
4 |
(3) |
14 |
27/11/12 to 31/1/13 |
7 |
2 |
(5) |
5. The period numbers are not drawn from statute or the evidence but are identifiers for the purposes of this decision. The dates and the lengths of the periods arise from the dates of the licences HMRC say that the Appellant held: the licences do not run from the beginning of a calendar month, and run for a period of calendar months selected by the person applying for them.
6. The assessment was not made period by period: that would have resulted in the use of higher monthly licence charges, but by identifying longer periods in which there had been, in HMRC’s view, a deficit of licences, and then applying the relevant “season ticket” rates to the longer periods. The Appellant, whilst it contested the deficiency, did not contest the method of calculation.
7. In preparing for the hearing HMRC had noticed that between 27 November 2009 and 6 January 2010 the Appellant had, according to their records, held four licences for B3 machines whereas the assessment had been made on the basis of only two for that period. As a result they concluded that the assessment was overstated by £860; they did not seek to defend the assessment to that extent. Table 1 shows the four licences now accepted by HMRC.
8. The issue in this appeal was thus whether the Appellant could satisfy us on the balance of probabilities that there was no deficit in the B3 licences held in any of the periods. It could do that in relation to an alleged deficit for any period either by showing that the number of licences held was greater or the number of B3 machines on site was smaller.
9. In its grounds of appeal that Appellant says that it would “produce site record cards and evidence from…B&N Regal which [will] refute HMRC’s calculations and how they have assessed AMLD category machines located on site.”
The Evidence
10. We heard oral evidence from Bruce Campbell, director of the Appellant, and from John Carpenter, director of B&N Regal.
11. Susan Gauld, an officer of HMRC, provided a witness statement in which she related the course of her enquiry and explained her conclusions. The Appellant did not contest the propositions of primary fact in that statement: what she did, who had said what to her when, and what she had seen. With the consent of Mr Ahmed we accepted those statements as evidence of what she had done, heard and seen, and did not hear oral evidence from her. We should say at this stage that it was clear that Miss Gauld had conducted a careful and balanced enquiry and had reached defensible conclusions.
12. We also had before us a volume of documents which included: correspondence between the parties, copies of invoices from the Appellant to B&N Regal, copies of some licences and applications for licences, copies of the Appellant’s machine site records, and print outs of returns made to the Gambling Commission by B&N Regal.
The Number of Licences
13. The evidence in relation to the number of licences granted in relation to the Appellant’s machines at B&N Regal was:
(1) the results of Miss Gauld’s interrogation of the HMRC database;
(2) the copies of licences which B&N Regal had found and sent to Miss Gauld during her enquiry;
(3) copies of applications for licences provided by HMRC’s Cumbernauld Accounting Centre and sent to Miss Gauld;
(4) the invoices from the Appellant to B&N Regal for the machines and for the licence duty paid;
(5) the evidence of Mr Carpenter in relation to the display of licences at New Coronet; and
(6) Mr Campbell’s evidence.
14. The evidence under headings (1) to (3) above was consistent with and supported the numbers of B3 licences in Table 1 above.
The invoices
15. The invoices showed that for each month in the period the Appellant invoiced B&N Regal separately for the supply of 14 machines and for the licence duty. For periods 1 to 6 it invoiced £2,250 per month for the machines and £1,708.88 for the licences. For periods 7 to 14 it invoiced £2,250 per month for the machines, and £1,835.31 for the licences.
16. We were not able to reconcile the charge made for the licences with either HMRC’s contention of the number of licences which were in force or with the Appellant’s figures for the numbers in each period of the different types of machine.
17. In Table 2, which follows, we set out the number of each type of machine which the Appellant contends were at B&N Regal (with a parenthetic comparison of the number of licences HMRC say were extant) and estimate the approximate yearly cost of such licences by reference to such information as to their cost as was in the papers before us. We then reduce that yearly cost to a monthly amount, and compare that with the sum invoiced.
Table 2
Period |
B3 machines per Appellant (no of licences per HMRC ) |
Annual cost per licence |
C machines per Appellant |
Total annual cost |
Approximate Monthly cost per licence £ |
Invoiced amount |
1 |
2 (2) |
2215 |
12 |
830 |
1,200 |
1,708 |
2 |
3 (4) |
2215 |
10 |
830 |
1,250 |
1,708 |
3 |
4 (4) |
2215 |
9 |
830 |
1,350 |
1,708 |
4 |
6 (6) |
2215 |
5 |
830 |
1,450 |
1,708 |
5 |
6 (6) |
2215 |
5 |
830 |
1,450 |
1,708 |
… |
… |
|
… |
|
… |
|
13 |
7 (4) |
2480 |
5 |
935 |
1,850 |
1,835 |
18. It will be seen that there is almost no correlation between the variations in the number of licences and their cost and the amount invoiced.
19. Mr Campbell told us that the recession in and after 2010 meant that it was not possible to recover increase in charges for licences (either because of increase in licence cost or number of licences); Mr Carpenter told us that the deal between them for the supply of machines had been informal – on a handshake, and that it had been a hard time economically.
20. As a result we find no support in the invoices for numbers of licences which differ from those which the earlier evidence indicates were in force. (Indeed the fact that in the early periods the cost of licences for the machines which the Appellant says were on site is less than the amount invoiced suggests to us that some of the consideration which would otherwise have been attributed to the VATable hire of the machines may have been attributed to the provision of the licences which was apparently treated as outside the scope of VAT.)
Mr Carpenter’s evidence
21. Mr Carpenter told us that he had received from the Appellant licences for the machines provided and had displayed them in the foyer at New Coronet. Mr Ahmed reminded us that a duty to display licences for machines on the premises was imposed by the Gaming Act and its breach involved a criminal sanction[1]: Mr Carpenter he said would thus have taken it very seriously. Mr Carpenter told us that the machines were labelled with their category so that a check against the licences was an easy matter. He told us that he made checks on a regular basis.
22. Although Mr Carpenter had not kept copies of all the licences he had had, he had been able to provide Miss Gauld with copies of 11 licences. None of these licences were not on the list which Miss Gauld extracted from HMRC’s records.
23. Mr Carpenter told us that when inspectors came from the Gaming Board they checked the licences. He said that there had been two inspections between 2007 and 2013: one in 2008 and one at some other time. The only problem the inspectors had found was that several of the licences had expired. This he said was found to be the result of a backlog in HMRC’s licence renewal system
24. We accept Mr Carpenter’s evidence that he displayed the licences he received from the Appellant. The problem he experienced with the expiry of licences suggested to us that he did not check the licences as diligently as he had remembered, and that it was possible that there had been periods when not all the necessary licences had been displayed.
Mr Campbell’s evidence
25. In a letter of 23 October 2013 to HMRC Mr Campbell says that he had noticed that towards the end of the AMLD regime (1 February 2013) the correct licences did not appear to have been in place, although he considered that they must have been applied for. It is not clear what he meant by “towards the end”.
26. Mr Campbell told us that he would have obtained licences for all the machines at New Coronet. He had not kept copies because there was no obligation to do so.
Discussion
27. The appellant provided no documentary evidence that licences other than those identified by Miss Gauld had been sought or granted. Although Mr Campbell said that payments for licences were made by direct debit no evidence of payments or reconciliation of payments to licences was offered. The licenses provided by Mr Carpenter matched Miss Gauld’s record, the licence applications provided by Mr Campbell revealed no missing items. Overall we are not persuaded that it was more likely than not that licences other than those found by Miss Gauld on HMRC’s system had been issued.
The Number of B3 Machines in any period
28. Miss Gauld prepared her assessments on the basis that for each of the relevant periods there had been 7 B3 machines at New Coronet. She had come to this conclusion because: (i) Mr Carpenter had told her on 8 May 2013 that prior to 1 February 2013 New Coronet had operated 7 B3 machines, (ii) Mr Carpenter confirmed this in a letter of 31 May 2013, (iii) on 12 August 2013 Mr Carpenter had told her in a telephone call that they had operated 7 B3 machines, (iv) on 14 August 2013 Mr Carpenter had said to her that in the previous three years (by which we assume meant after August 2010) he believed that they had operated 7 B3 machines, (v) the invoices from Arrow to B&N showed 14 machines divided between AWP and jackpot, (vi) in a telephone call with Mr Carpenter on 4 November 2013 Mr Carpenter had told her (apparently after consulting his son) that they believed that in the last four years (which we assume meant after about November 2009) they had 7 B3 machines.
29. We find the statements made to Miss Gauld clearly indicative that in the later periods there were 7 B3 machines at New Coronet, but not as strong evidence that that was the number in place before August 2010: remembering with any precision what was the case more than 3 years ago without any reference to written records or particular well remembered events is not in our view common.
30. The other evidence in relation to the number of B3 machines at New Coronet in the periods consisted of:
(1) the Appellant’s site records: sheets showing in relation to each position at New Coronet the machine occupying that position at a particular time;
(2) the invoices;
(3) Mr Campbell’s evidence; and
(4) Mr Carpenter’s evidence; and
(5) the Gambling Commission returns made by B&N Regal.
The site records
31. These records showed the number of machines in place at each of the 14 positions at New Coronet. They consist of sheets on the left hand side of which were details of the position and some details relating to the machine and on the other side names of machines followed by a date which we understood to be the date the machine was installed and on which it replaced its predecessor on the list (although it was not completely clear how this interacted with the machine details on the left hand side)
32. Mr Campbell had provided an analysis of these records to HMRC. The numbers of B3 machines in column 2 of Table 2 are drawn from that analysis.
33. HMRC pointed out that he had made a typographical mistake on one sheet of the analysis and made another mistake about the name of a particular machine. We accept that these were innocent mistakes which did not affect the truthfulness of his evidence, although they suggested that he was not the most accurate of recorders.
34. We compared the site records with the analysis produced by Mr Campbell. Table 3 shows the comparison. The entries in normal type are extracted from Mr Campbell’s site records; the emboldened entries “List says…” refer to the analysis produced by Mr Campbell. It will be seen that there are more than a few inaccuracies.
35. The site records were not so constructed as to make it easy or quick to extract an analysis of what machines were where at any time, and we found ourselves making a number of errors in our initial attempts. We concluded that the format of the records is likely to have contributed to the errors made by Mr Campbell, and indeed may also have made it difficult for him to know how many B3 licences he needed at any time.
36. It was also possible that there were errors in the site records: the errors made by Mr Campbell in extracting his analysis suggest that he was not a meticulous record keeper.
Table 3
|
4/09/09 to 26/11/09 |
27/11/09 to 6/1/10 |
7/1/10 to 29/3/10 |
30/3/10 to 12/4/10 |
1 |
Red Hot Roll |
Red Hot Roll |
Red Hot Roll |
Red Hot Roll |
2 |
Party Time |
Party Time |
Party Time |
Party Time |
3 |
Little Devils |
Little Devils |
Little Devils |
Little Devils |
4 |
Big Ben |
Skill Bingo |
Sierra Crane |
Sierra Crane |
5 |
Random Keys List says Rio Grand B3 |
Random Keys
B3 |
Random Keys
B3 |
Random Keys
B3 |
6 |
Mental Money Monster |
Mental Money Monster |
Mental Money Monster then Walk of Wealth from 10/2/10 List says Dream Factory |
Walk of Wealth List says Dream Factory |
7 |
Big Ben |
Good Vibrations |
Good Vibrations |
Good Vibrations |
8 |
Oceans 11 list says Emmarald Isle B3 |
Oceans 11 List says Egstra Lines
B3 |
Oceans 11List says Egstra Lines
B3 |
Oceans 11
B3 |
9 |
Crazy Fruits |
Crazy Fruits |
Crazy Fruits |
Elvis B3 |
10 |
Big Ben |
Nothing in position[2].List says Haunted House (B3) |
Haunted House B3 |
Haunted House B3 |
11 |
Empire State |
Empire State |
Empire State Until 23/1/10 then Happy Notes List says Happy Notes only |
Happy Notes |
12 |
Red Hot X |
Red Hot X |
Winning Ways B3 |
Winning Ways B3 |
13 |
Random Spinner |
Random Spinner |
Random Spinner Monkey Business from 3/3/10List says Random Spinner throughout. B3 |
Monkey Business B3 |
14 |
Bar X 5 |
Bar X 5 |
Bar X 5 |
Bar X 5 |
The invoices
37. Each of the invoices (after 30 April 2009) showed this description of what was provided:
“[12,13 or14] x awp machines and jackpot machines”
38. There was no indication of the number of B3 or machines of other categories. We did not find that these invoices assisted in determining the number of B3 machines in place at any time.
39. Mr Campbell told us that the number of B3 machines supplied to New Coronet had increased over time. They had gradually replaced C machines with B3 machines. That evidence was consistent with the site records and the number of licences shown on Miss Gauld’s interrogation of the HMRC licence computer.
Mr Carpenter’s evidence
40. We have noted Miss Gauld’s reports of what Mr Carpenter had told her, namely (i) that at 1 February 2013 B&N Regal had 7 B3 machines, and (ii) that in preceding years New Coronet had been supplied with 7 B3 machines.
41. Mr Carpenter had also written to Mr Campbell to confirm that (i) he had not taken advantage of the ability to have 8 B3 machines at New Coronet, and that greater numbers were introduced in later years, and (ii) that he had always had the correct number of AMLD licences displayed. He repeated that evidence to us.
The Gambling Commission returns
42. Each year to 30 June B&N Regal made a return to the Gambling Commission. This return required an answer to the question of how many of each category of machine there were on the premises. The question did not indicate a time at which the count should be made. The return showed the following:
43. Table 4
Period |
No of B3 Machines |
Gross Gambling Yield |
1 July 2009 to 30 June 2010 |
6 |
£93k |
1 July 2010 to 30 June 2011 |
7 |
£112k |
1 July 2011 to 30 June 2012 |
7 |
£156k |
1 July 2012 to 30 June 2013 |
8 |
£170k |
44. Mr Carpenter told us that in completing this for the first period in the table he had been aware that at the end of the period he had had 7 B3 machines but he had put 6 on the return as the average number of such machines in the period. His evidence was a little hazy about the nature of the entry saying initially that he did not know what date he took them from. We also noted that the named contact on the form was “Andrew”, not Mr Carpenter.
Discussion
45. As Table 4 shows, the Gambling Commission returns show the gross gambling income from the machines. Mr Campbell said that the effect of adding an extra machine of a particular class would not necessarily lead to a proportionate increase in income, since novelty use would spread between machines. We accept that. The Table indicates fluctuations in income and in mean income per machine. Over the whole period the income was about £18.5k per machine. That would suggest some 5 machines on average over the first year. Overall the return suggests to us that fewer than 7 B3 machines were on the premises in the year beginning 1 July 2009.
46. The time weighted average number of machines in that first year as appearing from the Appellant’s site records is about 3.5. Comparison with the Gambling Commission return therefore suggests that there were more B3 machines in place than the site records showed.
47. The rather broad brush approach of the Appellant to its invoicing B&N Regal, its less than meticulous approach to record keeping, and the analysis in Table 3 suggest to us that its site records may not be wholly accurate, but the Gaming Commission returns do not convince us that they are wholly wrong. We consider that Mr Carpenter’s statements to Miss Gauld about the early years may not have been wholly reliable and that his evidence does point to there having been some increase in the number of B3 machines over the period.
48. . We have not found coming to a conclusion an easy exercise, but on balance the evidence leads us to the conclusion that in periods 1 to 4 the Appellant had less than 7 B3 machines in place at New Coronet, and that the number of B3 machines grew over that period. We conclude that on balance the evidence shows that in periods 1 to 4 the Appellant had respectively 3,4,5,and 6 B3 machines in place and thereafter 7.
Conclusions
49. In our judgement the appeal succeeds in relation to period 1 to 4 to the extent that the deficit shown in Table 5 is less than that on which the assessment was based and otherwise fails.
Table 5
Period Number |
Dates |
Our conclusion as to the number of B3 machines |
No of B3 licences |
(Deficit) in licences |
1 |
|
3 |
2 |
(1) |
2 |
|
4 |
4 |
- |
3 |
|
5 |
4 |
(1) |
4 |
|
6 |
6 |
- |
5 |
|
7 |
6 |
(1) |
6 |
|
7 |
9 |
- |
Rights of Appeal
50. 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.