Loughshore Autos Limited v The Commissioners for HM Revenue and Customs: [2017] UKUT 0252 (TCC) [2017] UKUT 0252 (TCC)

UPPER TRIBUNAL
TAX AND CHANCERY CHAMBER
[2017] UKUT 0252 (TCC)
Between
LOUGHSHORE AUTOS LTDAppellantTHE COMMISSIONERS FOR HER MAJESTY’SRespondent
Before
JUDGE TIMOTHY HERRINGTONJUDGE CHARLES HELLIERDate 5 July 2017Category: Tax
[1]This is an appeal against a decision of the First- tier Tribunal 5 (the “FTT”) (Judge Staker and Mohammed Farooq) in which it dismissed an appeal by the Appellant against an assessment to excise duty under section 13(1A) Hydrocarbon Oil and Duties Act 1979 ("HODA").[2]The assessment had been made on the basis that between 28 August 2010 and 10 27 April 2012 some rebated kerosene had been used in the Appellant's coaches without an amount equal to the rebate having first been paid to HMRC.[3]The grounds of appeal can be grouped under two headings:(1) the first is that the Appellant argues that the assessment was made outside the time limits permitted by section 12A Finance Act 1994; 15(2) the second comprises a group of arguments that, broadly speaking, the FTT's conclusions were perverse: the Appellant argues that the FTT's decision erred in law on the grounds that (i) it misunderstood the evidence, (ii) wrongly considered that the assessment was a “best judgment” assessment, (iii) made a decision which so went against the weight of the evidence as to be perverse, and 20 (iv) was derived from a hearing in which the tribunal adopted a passive rather than an inquisitorial role thus limiting its ability to make just and fair findings. Factual Background[4]The following summary is of the relevant uncontentious findings of the FTT.[5]The Appellant operates a coach hire business. In pursuit of this business it lets 25 out coaches with its own drivers and also coaches for the hirer to drive. For some hires the customer is responsible for fuelling the coach (we shall call this type of hire as being on a "hirer-fuelling" basis, and hires when the Appellant is responsible for fuelling “Appellant-fuelling hires”).[6]On 28 April 2012 HMRC discovered that one of the Appellant’s coaches was 30 running on fuel which contained rebated kerosene. An investigation followed: at a visit to the Appellant's premises a further three of the coaches tested were found to be positive for rebated fuel. HMRC then sought further information from the Appellant and the Appellant responded. This information was provided on 7 September 2012 and included service records for fifteen coaches, miles per gallon figures for each 35 coach and details of fuel purchased by the Appellant.[7]On 18 October 2012 HMRC sent the Appellant a Notice of Intention indicating their intention to assess excise duty in relation to the period from 1 July 2009 to 27 April 2012, but inviting the Appellant to provide further information within 21 days. 3[8]The Notice set out a calculation of the duty which HMRC intended to assess based on the information supplied earlier by the Appellant. For each coach the miles travelled in the period were obtained from the service records, that was divided by the mpg figure for the coach given by the Appellant to give the volume of fuel used in the period. The total volume of fuel used was then compared with 5 the volume purchased as evidenced by the Appellant’s records. The excess of the former over the latter was treated as being rebated fuel on which a liability to duty was said to arise. The duty was computed as £10,611.[9]Within the 21 days period Mr Corr of Corr & Corr, the Appellant's accountant, 10 told HMRC that for some coach hires the customer was responsible for providing fuel. He explained that the company had two books in which hires were recorded and that one of them had perished in a fire; the other was supplied to HMRC.[10]On 12 February 2013, HMRC assessed the appellant to excise duty of £5,602. (The FTT noted that this was less than the amount of the 18 October 2012 15 computation "presumably to take account ... of occasions where the customer had been responsible for providing fuel").[11]Mr Corr then requested a review, raising again the situation in which the hirer was responsible for fuelling and the destruction of the second hire book in the fire. He also enclosed a letter from a customer of the Appellant, BCSL, indicating that it had 20 hired coaches on the hirer-fuelling basis.[12]On 10 April 2013, as a result of the review the assessment was withdrawn. The FTT understood that the main reason for the withdrawal was that the original assessment embraced the period from 1 July 2009 to 23 August 2010 but that on that latter date HMRC had tested the appellant's vehicles and found them to be free of 25 rebated fuel. On that basis, the assessment should cover the period from 24 August 2010 to 27 April 2012 (the "audit period") only. The review letter indicated that it was likely that a new assessment would be made.[13]HMRC started work on a new assessment. On 20 June 2013, they wrote to the Appellant noting that the earlier assessment had been based on fifteen vehicles and 30 asking whether there were other coaches.[14]On 29 June 2013, Mr Corr replied identifying another four coaches. HMRC carried out checks and concluded (in August 2013) that one of these coaches had been acquired after the audit period. They therefore excluded it from their computations.[15]On 9 December 2013 HMRC issued a new assessment for £12,717. The FTT 35 described that new assessment thus at [18] of its decision:
“...Most of the vehicles were owned by the Appellant for the whole of the assessment period, although some had only been acquired part way through that period, and this was taken into account in the assessment. The assessment based the mileage [of] each vehicle on information from vehicle service records and 40 information provided by the Appellant’s accountants. The assessment applied a miles per gallon figure for each vehicle based on information provided by the 4 Appellant or the Appellant’s accountants. A deduction from the fuel requirements was made where evidence was provided of a vehicle having been hired out on a self-drive basis. The calculation of the Appellant’s fuel requirements for all of its vehicles was considered to be 124,039.99 litres. The Appellant was found to have produced receipts for 5 102,051.04 litres during the revised assessment period, leaving a shortfall of 21,988.95 litres. The Appellant was assessed to duty on this shortfall.”
[16]Following a further review the assessment was reduced to £8,330. Ground 1: The Time Limit Issue 10 17. Section 12 HODA provides that no rebated oil shall be used as, or taken into a road vehicle, as fuel unless an amount equal to the rebate is first paid to HMRC. Section 13(1A) provides that on a contravention of section 12 HMRC may assess an amount equal to the rebate on any person who used the oil or was liable for the vehicle being taken on a road. Section 12A Finance Act 1994 (“FA 1994”) provides a 15 time limit as follows:
“(4) No assessment under [section 13 HODA] shall be made at any time after whichever is the earlier of the following times, that is to say— (a) subject to subsection (5) below, the end of the period of three years beginning with the time when his liability to the duty arose; and 20 (b) the end of the period of one year beginning with the day on which evidence of facts, sufficient in the opinion of the Commissioners to justify the making of the assessment, comes to their knowledge. (5) Subsection (4) above shall be without prejudice, where further evidence comes to the knowledge of the Commissioners at any time after the making 25 the assessment concerned, to the making of a further assessment within the period applicable by virtue of that subsection in relation to that further assessment. (6) The reference in subsection (4) above to the time when a person’s liability to a duty of excise arose are references – 30 (a) in the case of a duty of excise on goods, to the excise duty point; and (b) in any other case, to the time when the duty was charged.”
[18]Because they have been subject to judicial scrutiny to which we shall refer later it is convenient at this point to set out the corresponding limitation provisions in 35 section 73(6) Value Added Tax Act 1994 (“VATA”):
“(6) An assessment under subsection (1), (2) or (3) above of an amount of VAT due for any prescribed accounting period must be made within the time limits provided for in section 77 and shall not be made after the later of the following– (a) 2 years after the end of the prescribed accounting period; or 5 (b) one year after evidence of facts, sufficient in the opinion of the Commissioners to justify the making of the assessment, comes to their knowledge, but (subject to that section) where further such evidence comes to the Commissioners' knowledge after the making of an assessment under subsection (1), (2) or (3) above, 5 another assessment may be made under that subsection, in addition to any earlier assessment.”
The FTT's decision[19]At [55] the FTT rejected the argument that time began to run against HMRC in April 2012 when the inspection revealed that four coaches had been found to contain 10 some rebated fuel. It concluded that "in order to be able to raise an assessment relating to all 18 vehicles over the entire audit period, HMRC needed evidence of all those vehicles, and of their mileage and fuel consumption, as well as of the Appellant's legitimate fuel purchases over that period ..."[20]The FTT then recorded at [56] that the officer who raised the assessment under 15 appeal had said that "she had sufficient information to raise an assessment" on 18 October 2012 when the letter giving Notice had been sent, but the FTT said “However, the audit on which that letter and the subsequent 12 February 2013 assessment were based included only 15 vehicles, and was for a different audit period.” 20 21. After recording the receipt in June 2013 of the additional information in relation to the extra 4 vehicles, the FTT said:
“60. The Tribunal accepts that if HMRC had had for more than a year sufficient evidence to issue the assessment that it did in relation to 15 vehicles, then this new information in relation to a further three vehicles might not justify 25 the making of an assessment in relation to all 18 vehicles: see Cozens at [35] referred to in paragraph 49 above. The new information might justify the making of an assessment in relation to the three vehicles, but an assessment in relation to 18 vehicles would be invalid in respect of all of them: see Cozens generally. 30 61. However, the Tribunal accepts the argument in [ERF Ltd v HMRC [2012] UKUT 105 (TCC)] at [30], relied on by HMRC, that in relation to each of the vehicles, the assessment is not out of time merely because HMRC had sufficient information to make an assessment more than 12 months earlier, and that an assessment will only be out of time if HMRC had sufficient information 35 to make the assessment that it did in relation to that vehicle more than 12 months earlier. [original italics]”
[22]Having explained that HMRC’s computation was dependent upon the numbers of coaches operated in the audit period, the FTT then concluded that the time limit under section 12A (4) (b) did not start to run until, at the earliest, 29 June 2013 when 40 the Appellant provided details of the additional four vehicles. The parties’ submissions 6[23]Mr McNamee argues, perhaps with some justification, that it is difficult to reconcile paragraphs [60] and [61] of the FTT's decision. He says that the assessment under appeal was a single global assessment and, relying upon Pegasus Birds v CCE [1999] STC 95 and [2000] STC 91, and John Cozens v HMRC [2015] UK FTT 482 (TC) says that if any element of such an assessment is out of 5 time then the whole assessment is out of time.[24]He says that the global assessment was in respect of each excise duty point which had arisen in the audit period. Those excise duty points arose when rebated oil was taken on board a coach or a coach was driven on a road. They related to 10 individual coaches and specific times even though it could not be said when those times were. The assessment under appeal comprised excise duty points in respect of oil being taken on to 18 coaches. HMRC had, on the officer’s evidence, sufficient information to raise an assessment, and indeed had raised an assessment, in relation to the excise duty points relating to the 15 coaches of which they knew on 18 October 15 2012 when they wrote their letter giving notice of their intention to assess. And the assessment under appeal was made more than 1 year after 18 October 2012.[25]Thus, he says that each element of the assessment which related to those 15 coaches was out of time, and as a result the assessment under appeal was out of time.[26]Mr McNamee says that it would emasculate the intended protection of the 20 limitation period in s 12A(6)(b) FA 1994 to permit the clock to be reset each time HMRC received any material which could result in a different assessment.[27]Mr Charles puts his case in the alternative. First in support of the FTT’s reasoning, he submits that while HMRC could have raised an assessment on the basis of 15 vehicles, they could not have raised the assessment under appeal until they were 25 told on 29 June 2013 of the additional three vehicles. That assessment was therefore not out of time. He argues that the FTT was not wrong to rely upon paragraph [30] of EDF.[28]In the alternative Mr Charles argues that it was not until 19 December 2012, when the Appellant supplied details of its coaches and its hire book to HMRC, that 30 HMRC had sufficient information to raise an assessment, and that the assessment under appeal was made within 12 months of that date. Discussion[29]The relevant words of s 12(4) (b) FA 1994 match those of s 73(6)(b) VATA. It is clear, and was not disputed, that the authorities relevant to the construction of the 35 latter provision are relevant to the former.[30]The principles to be applied in interpreting section 73(6)(b) were set out by Dyson J in the High Court in Pegasus Birds at pages 101g to 102 as follows: “1. The commissioners' opinion referred to in s 73(6)(b) is an opinion as to whether they have evidence of facts sufficient to justify making the assessment. 40 Evidence is the means by which the facts are proved. 7[2]The evidence in question must be sufficient to justify the making of the assessment in question (see Customs and Excise Comrs v Post Office [1995] STC 749 at 754 per Potts J).[3]The knowledge referred to in s 73(6)(b) is actual, and not constructive knowledge (see Customs and Excise Comrs v Post 5 Office [1995] STC 749 at 755). In this context, I understand constructive knowledge to mean knowledge of evidence which the commissioners do not in fact have, but which they could and would have if they had taken the necessary steps to acquire it.[4]The correct approach for a tribunal to adopt is(i) to decide what were the facts 10 which, in the opinion of the officer making the assessment on behalf of the commissioners, justified the making of the assessment, and(ii) to determine when the last piece of evidence of these facts of sufficient weight to justify making the assessment was communicated to the commissioners. The period of one year runs from the date in (ii) (see Heyfordian Travel Ltd v Customs and 15 Excise Comrs [1979] VATTR 139 at 151, and Classicmoor Ltd v Customs and Excise Comrs[1995] V&DR 1 at 10).[5]An officer's decision that the evidence of which he has knowledge is insufficient to justify making an assessment, and accordingly, his failure to make an earlier assessment, can only be challenged on Wednesbury principles, or 20 principles analogous to Wednesbury (see Associated Provincial Picture Houses Ltd v Wednesbury Corp [1948] 1 KB 223) (see Classicmoor Ltd v Customs and Excise Comrs[1995] 14 V&DR 1 at 10–11, and more generally John Dee Ltd v Customs and Excise Comrs [1995] STC 941 at 952 per Neill LJ).[6]The burden is on the taxpayer to show that the assessment was made outside 25 the time limit specified in s 73(6)(b) of the 1994 Act.”[31]The case went on to the Court of Appeal which dismissed the appeal for the reasons given by Dyson J but without expressly approving these formulae.[32]In ERF (referred to at [61] of the FTT’s decision) the appellant’s accountants had provided HMRC successively with three reports estimating the VAT which had 30 been lost by reason of the dishonesty of a member of the appellant’s staff. These were called BDO/1 to BDO/3. The appellant argued that an assessment was out of time with regard to section 73(6)(b) because there had been two prior times at which HMRC could have raised an assessment and the failure to make such an assessment was Wednesbury unreasonable. Before turning to this argument, the Upper Tribunal 35 said that it was important to note one important point about the way the case was put: “30. At one level it might be thought that Dyson J’s principles applied in relation to the actual assessment made in the case in hand. On this footing the assessment referred to in principle 5 would be the actual assessment ultimately made, not an earlier, and different, assessment. If this were the approach then there would be a 40 short answer to the present case. Any actual assessment has to be based on properly constituted figures for the periods in question. The actual assessment in the present case was based on figures that only became apparent as a result of BDO/3, and contained the figures from that document (save for one period). For almost all of the periods the figures were different from those which had 8 appeared in previous reports, and no-one suggested that the actual figures that were assessed could have been reached any earlier than BDO/3. That being the case, it is plain that HMRC could not have raised these assessments earlier than the date of receipt of that document, so it would not be possible to mount a challenge under principle 5 on the footing that a decision 5 not to raise that assessment was unreasonable before receipt of the report.[31]However, Mr Harris’s challenge is not based on that premise. His challenge is based on the premise that a different assessment could and should have been made earlier, on one or other of the two occasions mentioned above. On those 10 occasions HMRC had different quantifying information available to it in the form of the earlier reports, and the earlier assessments should have been the basis of an assessment which would, could and should have reflected those figures. On each of those occasions the jigsaw referred to in Pegasus principle 4 was complete. 15 32. The tribunal below said at paragraph 100 that the only “relevant question was whether the assessments had been made within one year after evidence of facts, sufficient in the opinion of the Commissioners to justify the making of the assessment, came to their knowledge”. Miss Shaw said this was the right question. The Tribunal below addressed Mr Harris’s analysis. This is apparent 20 from the Tribunal’s paragraph 112. “112. We now turn to consider whether, as Mr Harris argued, there was an earlier time at which Customs had sufficient evidence on which an assessment could be justified, and whether it was wholly unreasonable or perverse for Customs at any of those times to have delayed the making of 25 the assessment, so that if the earlier assessment was made more than one year after that time it would be out of time. Mr Harris submitted that there were a number of occasions on which Customs had sufficient evidence to justify the making of an assessment. We consider each of those submissions in turn.” 30 HMRC through Miss Shaw accepted that this again was the correct question. If HMRC succeed on Mr Harris’s approach they plainly succeed on this part of the appeal. We therefore deal with this point, and do not consider whether Dyson J’s point 5 requires a different approach.”[33]It is clear from these passages that the Upper Tribunal did not decide in ERF 35 that a limitation challenge could succeed against an assessment only if it would have been Wednesbury unreasonable not to have made the very same assessment earlier. HMRC’s case was not put that way, and the Upper Tribunal made no decision in relation to that argument.[34]Paragraph 61 of the FTT's decision appears to accept the argument articulated 40 but not decided upon in ERF at [30]. The juxtaposition of paragraphs [60] and [61] and the FTT's decision indicates that the FTT considers that this principle overrode that in Cozens.[35]In Cozens the tribunal said: 9 “29. It is well established in respect of VAT assessments that if the assessment concerned is to be construed as being a single, global assessment covering a number of different prescribed accounting periods rather than as being a series of separate assessments in respect of each of the prescribed accounting periods then the whole assessment will be invalid on grounds of being out 5 of time if any one of its elements is shown by the taxpayer to be out of time. 30.This principle was established in the Court of Appeal’s judgment in S J Grange Limited v CEE [1979] STC 183. Lord Denning MR, having found that there was no reason why a single assessment could not be made covering a 10 number of prescribed accounting periods rather than having to issue separate assessments in respect of each period, [said] that the long stop date for an assessment now contained in s 73 (6) (a) VATA was to be measured by reference to the end of the first prescribed accounting period included in the assessment: see page 193 of the judgment. He said that in all cases where it is 15 impossible for the commissioners to split the assessment up into the prescribed accounting periods (for example where the Commissioners were unable to say with certainty which accounting period the liability fell within) the Commissioners could assess for any period of time they specified covering multiple accounting periods. 20 31.Templeman LJ elaborated the test a little more at page 195 when dealing with the objection that if a global assessment were permitted it could lead to stale demands being included in the assessment as follows:
“There are two ways of dealing with that. The first way is to do what counsel for the company would have us do, and that is to make us do 25 violence to sub-s (1). The second way is to fit the expression 'prescribed accounting period' used in sub-s (2) to the actual periods covered by the assessment which has been made under sub-s (1). If the latter course is followed, far from this being disadvantageous to the company, it means that the limitation period starts with the very earliest prescribed accounting 30 period of three months which is covered by the assessment which has been made on him, so that if the commissioners take that course under no circumstance can the company be prejudiced.”
[36]That all related to the time limit in section 73(6)(a). Then at [35] it derived from the judgment of Woolf J in International Language Centre Ltd v CCE [1983] 35 STC 394 at 396 the conclusion that the principle in Grange was equally applicable where the relevant time limit was that in section 73(6)(b). It recapitulated this conclusion at [123] where, in rejecting submissions from HMRC that if the assessments were in a single sum the time limit question would be answered by establishing when HMRC received the last piece of the 40 jigsaw, it said: “123... The principle that if part of a global assessment is out of time the whole assessment fails was clearly established to protect the taxpayer against the prejudice that could be caused to him by [HMRC] choosing to use a global assessment rather than a series of separate assessments and 45 thereby bundling together demands that would be out of time if made separately with those that were in time ... the time limit provisions are there to protect the taxpayer from tardy assessment; permitting [HMRC] to 10 use a global assessment in the way envisaged by [counsel for HMRC] to defeat assessments in respect of particular excise duty points that would otherwise be out of time is inconsistent with that principle.[124]In my view the VAT authorities demonstrate that when a global assessment is made, so as to create an accounting 5 period which covers a number of prescribed accounting periods, then for the assessment to be valid it must be in time for all of the prescribed accounting periods that it covers. That is readily apparent from the way the Court of Appeal formulated the principle in Grange ... where the statutory wording was 10 interpreted so as to read in words permitting an assessment to be expressed to cover a period including a number of prescribed accounting periods.[125]There is nothing to suggest that this reasoning is confined to assessments falling within what is now section 73(6)(a); the introductory words in section 73(6) which covers both (a) and (b) makes it clear that 15 whether VAT is due is calculated by reference to prescribed accounting periods. I therefore find the reasoning of Woolf J in International Language Centres and Simon Brown J in Spillane, which clearly assumes the reasoning in Grange covers both provisions, highly persuasive.”[37]There is nothing in ERF which contradicts these conclusions. We agree with 20 them. At [127] of Cozens the tribunal held that the same approach should be taken in respect of excise duty assessments. The tribunal noted that s 12(4) FA 1994, which sets out the relevant time limit for excise duty assessments, provided that the time limit applied to “an assessment of the amount of any duty of excise due from any person” and that time began to run when a person’s liability to the duty arose. The 25 tribunal held that the effect of this provision when read with s 12(6) FA 1994, which stated that the reference in s 12(4) to the time when a person’s liability to excise duty arose was a reference to the excise duty point of the goods in question, was that where an assessment was made covering a number of excise duty points arising during the period the assessment was stated to cover, then in order for the assessment to be valid 30 it must be in time with respect to all of the excise duty points to which it relates.[38]We agree with that reasoning and that it can be applied to assessments made pursuant to s 13(1A) HODA for the following reasons.[39]In relation to assessments in respect of improperly used rebated fuel, as in this case, the effect of s 13(1A) HODA is that the liability to the duty rises at the time the 35 fuel is taken into the vehicle in question. As with s 12(4) FA 1994 in relation to excise duty assessments, s 12A(4) FA 1994 provides that time begins to run when a person’s liability to the duty arose so that the effect of this provision when read together with s 13(1A) HODA is that time begins to run when the vehicle in question is fuelled, that being the equivalent of an excise duty point (although s 13(1A) HODA does not 40 contain an equivalent to section 12(6) FA 1994, it is clear that the liability to the duty arises on the contravention, and making of the assessment is part of the machinery for its collection). Consequently, as with excise duty assessments, where an assessment is made, as in this case, covering a number of occasions when a vehicle has been filled with rebated fuel during the period the assessment was stated to cover, then in order 45 for the assessment to be valid it must be in time with respect to all of the contraventions to which it relates. 11[40]The same reasoning applies to assessments falling within s 12A (4) (b) FA 1994: see Cozens at [125], quoted at [36] above, the relevant wording of this provision being identical to that in s 12(4) (b) FA 1994.[41]There is a difference between the VAT provisions and the excise duty provisions on which we should comment. The VAT provisions 5 state that no assessment may be made after the later of the periods mentioned in sub-paragraphs (a) and (b) whereas the excise duty provisions proscribe the making of an assessment after the earlier of corresponding periods in sub-paragraphs (a) and (b). Thus, the excise duty provisions have a three-year long stop which is not available in a VAT 10 context. Nevertheless, the similarity of the drafting of the relevant provisions in the evident object of protecting the taxpayer from tardy assessments convinces us that the same construction should be applied to both.[42]Consequently, the correct approach for a tribunal when faced with a global assessment of this type and considering the application of s 12A (4) (b) FA 1994 is to 15 ask itself two questions:(1) was it perverse of HMRC not to have made an assessment at any time earlier than the time when the assessment was actually made? and(2) if the first question is answered in the negative, was the assessment in time in relation to all of the contraventions of s 12 HODA? 20 43. If therefore in relation to any contravention of s 12 HODA to which the assessment relates there was a time when it would have been perverse not to make an assessment (when an assessment “should” have been made) by reference to evidence available at that time, a later actual assessment cannot escape the limitation of section 12A by bundling into the computation of that later assessment facts by reference to 25 which the earlier assessment “should” have been made. Accordingly, a global assessment will be proscribed by section 12A if such is its effect.[44]Therefore, when the FTT concluded that an assessment “would only be out of time if HMRC had sufficient information to make the assessment that it did in relation to that vehicle more than 12 months earlier” it erred in law. 30 45. However, in our judgment that does not affect its conclusion.[46]We turn to the first of the two questions set out at [42] above, that is whether it would have been perverse for the officer not to have made an assessment at some time more than a year before 9 December 2013 when the assessment under the appeal was made. 35 47. The question is whether it was perverse not to assess, not whether it was or was not perverse to consider that she had sufficient information to assess. There will be cases where HMRC have sufficient information to justify an assessment but seek further information to justify a better assessment. In such a case if it was not unreasonable to seek that further information it will not have been perverse to decide 40 not to assess even if an assessment could have been made at the earlier time. 12[48]That was the approach of the FTT in ERF. In that case, there was a time when the relevant officer knew of the dishonesty which was necessary to found an assessment and had received a report with figures which would have enabled him to make an assessment but he agreed that the appellant's accountants should revise and improve that report. The FTT in that case found that the officer’s initial 5 willingness to accept that report was overtaken by his actual acceptance of the further investigation. It said that it was not enough that there be merely evidence of facts. The evidence must be sufficient, taking into account the obligation to exercise best judgment, to justify the making of the assessment. By accepting the further investigation, the 10 officer was accepting that the first report did not have the necessary quality of factual evidence on which an assessment could (the Upper Tribunal said “should”) be based.[49]The Upper Tribunal at [44 and 45] found that that conclusion was open to the tribunal: the officer had held off from forming a final view until the taxpayer had done more work and a final view was not formed. That was not an unreasonable view 15 for HMRC to take.[50]In this case, it was not in our view perverse not to assess merely on the basis of the tests which had revealed that four coaches had rebated fuel in their tanks. Until the receipt of the additional information on 7 September 2012 all HMRC had was a broad understanding of the Appellant’s business and the knowledge that four coaches had 20 tested positive for rebated fuel. It was perfectly reasonable for HMRC to seek better to quantify the excise duty which might have been lost before raising an assessment.[51]On 7 September 2012 HMRC received further information from the Appellant. The processing of that information led to the letter of 18 October. In our view, it was perfectly reasonable for the officer to await the processing of that information, the 25 preparation of the calculations and the seeking of comments from the Appellant before forming a review that she had sufficient information to make an assessment.[52]The next time for consideration is 18 October 2012 when HMRC wrote to the Appellant giving notice of its intention to assess, set out basis on which that assessment would be made and its amount, but invited the Appellant to provide any 30 further relevant information within 21 days. It was also at that point that the officer said in her evidence to the FTT that she considered that she had sufficient information to issue an assessment.[53]In our view, it was not perverse for the officer not to make an assessment at the time of the Notice, and reasonable for her to wait until either the Appellant responded 35 to the invitation to provide further information or the 21 days elapsed. The officer asked if the Appellant wished to make representations. By accepting the possibility that those further representations would improve the quantum and quality of the evidence to justify the making of an assessment the officer was not acting perversely.[54]The next relevant time is on the receipt of the additional information – with Mr 40 Corr’s letter about hirer-fuelling - shortly after 19 December 2012. At that time HMRC had all the evidence it needed to make assessments by reference to the 15 coaches. 13[55]We therefore answer the first question by concluding that there was no time more than one year before 9 December 2013 when it would have been perverse for HMRC not to have assessed.[56]We now turn to the second question. As we have stated at [54] above, it was not until 19 December 2012 that HMRC had the last piece of evidence 5 that would justify it making an assessment in relation to the 15 coaches. An assessment therefore needed to be made within 12 months of that date for it to be in time in relation to all of the contraventions relating to those coaches. The assessment was actually made on 9 December 2013 which was less than 12 months after 19 December 2012 when the 10 assessment by reference to all 18 coaches was made. As a result, the assessment is not out of time by reference to the information received on 19 December 2012.[57]In relation to the remaining coaches, the last piece of evidence that would justify the making of the assessment was not received until 29 June 2013. Accordingly, the assessment was also in time in relation to the contraventions relating 15 to those coaches.[58]We therefore distinguish the position in this case from that in Cozens. In that case, the global assessment was held to be out of time because in relation to a number of the excise duty points relating to goods which were the subject of the assessment HMRC, on their own admission, had evidence sufficient to justify the assessment in 20 relation to those goods more than 12 months before the global assessment was made: see [128] of the decision.[59]We therefore conclude on the time limit issue that there was no time more than a year before 9 December 2013 when it would have been perverse for HMRC not to have assessed. The assessment was made less than 12 months after the receipt of Mr 25 Corr’s letter of 19 December 2012 and less than 12 months after the receipt of his letter of 29 June 2013. It was therefore not made outside the time limit in section 12A.[60]Consequently, for different reasons than those given by the FTT, we dismiss the appeal on the Time Limit Issue. Ground 2: The decision was perverse 30 61. As set out at [3 (2)] above, the Appellant relies on four arguments to support its contention that the FTT’s conclusions were perverse and accordingly involved the making of errors of law. We shall deal with each of the four arguments in turn. The FTT misunderstood Mr Corr’s evidence[62]There are two limbs to this criticism. The first is that the FTT’s decision 35 rejected Mr Corr’s evidence on the grounds that he had no first -hand knowledge of the business and that this reflected a misapprehension of the nature of his evidence: The second limb relates to an apparent confusion in the FTT’s decision as to the two types of hire. 14 The first limb[63]Mr McNamee says that Mr Corr gave evidence of the accounting exercise he conducted as an independent expert, giving evidence of a purely accounting nature. The important aspect of Mr Corr’s evidence was that he could identify in the bank statements those hires which were hirer-fuelling. That had 5 not been taken into account. The way in which the FTT chose to give no weight to this evidence displayed a misunderstanding of it.[64]We start by describing the nature of that evidence and how the FTT dealt with it. 10 65. We have described the methodology which HMRC used to produce an indicative figure in the Notice of Intention, the 12 February 2013 assessment and the 9 December 2013 assessment. In each case, from information supplied by the appellant, the officer had:(1) calculated the total miles travelled by each of vehicles in the audit period; 15(2) thence calculated the mean daily mileage;(3) estimated the appellant-fuelling miles travelled in each duty period (the period during which a particular rate of duty applied in the audit period);(4) made an adjustment for a hirer-fuelling in making the 12 February 2012 assessment and 9 December 2013 assessment but only to the extent that such 20 hire appeared in the first hire book;(5) estimated the volume of fuel which each vehicle would have consumed in each period in relation to appellant-fuelling hires, being the estimated appellantfuelling mileage for the period divided by the miles per gallon figure for the vehicle supplied by the appellant; 25(6) thus estimated the total volume of fuel consumed by the coaches in the period. (VolC)(7) calculated the total volume of fuel purchased in the period (VolP) from records of the appellant’s fuel purchases.[66]The excess of VolC over VolP was treated as fuel which should have borne full 30 duty but did not (the "lost fuel" as it was called); and the assessment was made by multiplying the excess for each duty period by the duty rate for that period and aggregating the results.[67]The thrust of Mr Corr’s evidence before the FTT was that the adjustment made at step (4) was too small. The adjustments made by HMRC in relation to the 35 assessments reflected only entries in the rental book which had been supplied to them but no adjustment had been made for hirer-fuelling hires in the book which Mr Corr said had been destroyed in the fire. If there had been more hirer-fuelling days there would have been fewer appellant-fuelling days and VolC would have been lower. As a result, the assessment should have been smaller. 15[68]Mr Corr had undertaken an exercise to attempt to quantify the effect of the omission. In this exercise, he produced a calculation comparing:(1) an estimate of what hirer-fuelling income would have arisen on the basis that the "lost fuel” as calculated by HMRC related only to hirer-fuelling hires 5 (X); with(2) figures he extracted from the Appellant's bank account which he regarded as representing the actual hirer-fuelling takings which were not in the first hire book (Y).[69]This comparison showed that for each duty period between 73% and 100% of 10 the lost fuel could be treated as represented by such hirer-fuelling takings.[70]Mr Corr had estimated X by:(1) calculating the mean daily lost fuel using HMRC's calculations;(2) using that and HMRC’s calculation of vehicle mileage in each duty period to calculate an estimate of lost fuel per duty period; 15(3) using that and a figure of 20 miles per gallon to estimate the "lost miles" per duty period;(4) using that to estimate, on the basis of 360 miles per such hire and a hirerfuelling fee of £500, what income would have resulted from the "lost fuel".[71]Mr Corr’s figure of Y was, the FTT recorded, obtained from bank statements by 20 isolating those receipts which(i) were for a small amount (since he said that hirerfuelling hire tended to be for smaller amounts), and(ii) were not related to hires shown in the first hire book.[72]The FTT accorded no weight to Mr Corr’s evidence. It found at [74] that:
"the Appellant has not produced evidence capable of showing that the HMRC 25 figure is wrong, and of showing positively what corrections should be made in order to make the assessment right or more nearly right."
[73]If the FTT had accepted Mr Corr’s analysis of the Appellant’s bank statements, the logical consequence would have been that HMRC’s estimate for VolC would have been too high and the assessment too large. By its conclusion that the Appellant had 30 not produced evidence showing what corrections should be made the FTT rejected the evidence of Mr Corr’s analysis of the bank statements.[74]The FTT set out its conclusions on Mr Corr’s evidence at [71] and 73]:
“71. Evidence was given by Mr Corr. However, Mr Corr is the Appellant’s accountant. He said in his evidence that he had no involvement in the 35 Appellant’s business, and it was evident that he had no first-hand knowledge of how the Appellant’s business operates. It would appear that he has no first-hand knowledge of the claim that the Appellant had two rental books that it was operating during the audit period, one kept in the vehicles and one kept in the 16 business premises. His evidence to that effect was apparently based on what his client had told him.”
[75]This is not a criticism of the mechanics of the exercise Mr Corr conducted but of his ability to give weighty evidence as to the accuracy of the assumptions used in his calculations. Unless his assumptions were right, his calculations 5 could be of no probative value.[76]Paragraph [71] continues:
“…In any event, even if it was true that there were two books and that one of them had been lost in a fire, it cannot be known what information was in the lost 10 book about occasions on which customers were responsible for providing fuel. In the absence of the second rental book, the burden would still remain on the Appellant to establish by some other means a more reliable figure than that used by HMRC.”
[77]This passage appears to disregard the possibility that “some other evidence” 15 might indeed be Mr Corr’s evidence. For, given that no criticism is made of his accounting expertise, if it was shown that hirer-fuelling hire was for smaller amounts his evidence was a means of estimating a more reliable figure.[78]Then, at [73], the FTT returns to Mr Corr’s evidence: “73. The Appellant relies on calculations prepared by Mr Corr. The calculations 20 in his schedule were based on a typical hire journey of 360 miles and a typical hire cost of £660 for such a journey. However, if Mr Corr had no involvement in the Appellant’s business, it is not apparent on what basis he could give evidence that these figures were indeed typical. In his oral evidence, he admitted that they could be described as “arbitrary” figures….” 25 79. This part of this paragraph addresses the assumptions used in relation to Mr Corr’s estimation of X; the points made are irrelevant to the calculation of Y. It is only in the calculation of X that £500 and 360 miles are relevant.[80]The FTT continues at [73]: “For the hearing, there was a more detailed schedule of lodgements produced by 30 Mr Corr indicating which lodgements into the Appellant’s bank account related to hires where the customer was responsible for providing fuel. However, it was entirely unclear on what basis Mr Corr could reliably give any such indication.[81]It is here that the FTT explains why it disregards Mr Corr’s evidence in relation to the calculation of Y. 35 82. Earlier in its decision at [34] the FTT had explained Mr Corr’s methodology and at [39] recorded his answers to a number of cross examination questions about the recording and documentation of hirer-fuelling hires. There it noted that Mr Corr did not know whether the hirer-fuelling hire contracts for the hire of a coach without a driver provided for the hirer to fuel; and that Mr Corr was not involved in the business 40 and could not know for certain that the smaller sums he had extracted related only to hirer-fuelling contracts. 17[83]Thus, it seems to us that the reason the FTT rejected Mr Corr’s evidence was because it was insufficiently robust to enable it to conclude that it was more likely than not that the assumptions which Mr Corr had made about hirer-fuelling hires (and which he had applied in calculating Y), were correct, and that there was no other evidence before the tribunal that 5 this was the case.[84]This was not a rejection of Mr Corr’s accounting expertise or of the accounting exercise undertaken by him, but a conclusion that there was insufficient evidence that his assumptions were correct. This does not to our minds display a misunderstanding of Mr Corr’s evidence. We address later whether it was perverse of the tribunal to 10 conclude that there was insufficient evidence to reject those assumptions. The second limb[85]Mr McNamee contends that the FTT’s decision appears to display a confusion between the two types of hire. That confusion contributes to a submission that it misunderstood the evidence and so erred in law. 15 86. The FTT used a number of different terms:
“self-drive”, “self-hire” and “coach only hire”
. These terms were used: (1) At [18], where the FTT say that in the computation of the fuel requirements “[a] deduction from the fuel requirement was made where evidence was provided of a vehicle having been hired out on a self-drive basis. 20 The implication here is that self-drive hire was hirer-fuelling hire; (2) At [26], where the FTT says that “McNamee McDonnell Duffy Solicitors, submitted a Schedule created by Mr Corr, showing the amount of monies generated by the Appellant’s self-hire business, and the average mileage of such journeys and the amount of fuel required”. 25 This appears to treat self-hire hire as being the same as self-drive hire, and that as hirer-fuelling hire; (3) At [34] in describing Mr Corr’s evidence of how he extracted sums from the bank statements:
“Vehicle rentals on a self-hire basis tended to be to large bodies who paid every two weeks or once a month. On the other hand, a small 30 lodgement of £100 would probably be for a private party hiring a vehicle on an owner operated basis. Mr Corr produced for the hearing a schedule of all lodgements during the audit period, and indicating which of the lodgements he believe related to private hire.”
. This appears to adopt the opposite meaning of self-hire hire to that used in the 35 earlier paragraphs. Here self-hire seems to mean appellant fuelling hire. The use of “owner operated” is confusing; (4) At [38] in further description of Mr Corr’s evidence:
“Mr Corr further confirmed that HMRC had given credit for each occasion on which the Appellant had provided evidence that the vehicle had been hired out on a self40 drive basis. He confirmed that the Appellant’s case was that there were additional occasions on which the vehicles had been hired out on a self-drive 18 basis for which credit had not been given by HMRC, due to the fact that the evidence had been lost in a fire.”
. Here self-drive hire is used in the same way as in (1) above; (5) At [39]:
“He had said in examination in chief that customers hiring out vehicles on a self-drive basis tended to pay fortnightly 5 or monthly, so that payments for self-drive rentals tended to be larger amounts, while lodgements for rentals on an owner operated basis tended to be for smaller amounts. However, Mr Corr’s schedule of lodgements showed some smaller amounts as being for self-drive rentals, yet did not include some larger amounts as self10 drive rentals. It was also noted that some of the larger payments indicated as self-drive rentals were for odd amounts, rather than rounded figures. “. Here the FTT uses self-drive hire as having the opposite meaning to that it takes in (1) and (3) above, and owner operated to have the opposite meaning to that used in (3); 15 (6) At [47]: “HMRC’s officer “gave credit to the Appellant where she had documentary evidence of hire on a self-drive basis”
. Here self-drive is used consistently with its use in (1) above,[87]Later, at [72] the FTT said:
“Evidence might also have been given by customers of the Appellant who provided fuel themselves when hiring vehicles. The Tribunal 20 was told that some of these customers were bodies such as education or health authorities, who presumably would have retained their own records of such matters.”
. This, says Mr McNamee is confused or wholly wrong. The evidence was clear: customers such as Local Authorities did not provide their own fuel.[88]We accept that there is some confused narration in the decision, but read as a 25 whole the meaning is clear. The important points which were clearly understood by the FTT were:(i) that there was a category of hire in which the hirer paid for its own fuel,(ii) Mr Corr’s evidence was that persons in that category paid small irregular sums, and(iii) from that Mr Corr sought to identify what receipts related to hirerfuelling hires. 30 89. Although the muddle is unfortunate it does not display a failure to understand the nature of Mr Corr’s evidence. The FTT wrongly considered that the assessment was a “best judgment” assessment[90]At [65] the FTT, citing Thomas Corneill v HMRC [2007] EWHC 715 (Ch), said that HMRC were “entitled to make a best judgment assessment” and were entitled to 35 do so using the methodology employed.[91]In Mr McNamee’s skeleton argument for this tribunal he argues that the assessment was a matter of arithmetic being applied to parameters which had been supplied by the Appellant and agreed between the parties rather than of best judgment. Before us he said that the FTT had erred by placing too much weight on 40 best judgment. 19[92]Section 73 VATA speaks of HMRC making an assessment to “the best of their judgment” whereas the assessment provision in section 13(1A) HODA says only that where oil is applied in contravention of section 12, “the Commissioners may assess” an amount equal to the duty.[93]In Thomas Corneill Mann J said at [32-33] that nothing could 5 be read into the absence of a reference to “best judgment” in section 13 HODA because some judgment was necessary to make almost any assessment. That is not authority for regarding section 13 as providing that an assessment should be made to best judgment as the FTT’s quoted passage may suggest. 1094. In our view the FTT’s reference to best judgment is unfortunate but does not obscure the nature of their finding, namely that they regarded HMRC as entitled to make an assessment. There is nothing in the statutory provisions which proscribes the making of an assessment on the basis of arithmetical calculations performed on largely agreed figures. 1595. But the nub of Mr McNamee’s point is that the FTT appeared confused about the nature of the assessment and as a result might have been equally confused about the nature of the evidence given by Mr Corr. It seems to us however that the two are unrelated: the first relates to the operation of a statutory provision, the second to the weight given to particular evidence. We do not find that the apparent confusion casts 20 doubt on the FTT’s assessment of the evidence. This may have been an error of law but it was not an error material to the FTT’s decision. The decision was against the weight of the evidence[96]Mr McNamee says that the FTT’s decision, which was effectively that there were no hirer-fuelling hires other than those in the first book, went against the weight 25 of the evidence and was perverse:(1) It was irrational to give no weight to any of Mr Corr’s evidence;(2) It ignored the letter from BCSL(3) It ignored the fact that the other information on which HMRC had made their calculations used evidence given to them by Mr Corr; 30(4) It ignored the fact that the “lost fuel” as calculated by HMRC was only some 2% of the Appellant’s fuel consumption and that the result of the calculation was a figure which properly should be regarded as lost in the roundings and estimates; and(5) It ignored the unreliability of HMRC’s calculations which had given rise 35 to four different figures.[97]We take these in turn. 20 Mr Corr’s evidence[98]The only oral evidence the FTT received was that of Mr Corr. The FTT commented on the lack of other evidence at [72] as follows:
“Even if one of the rental books was lost in a fire, the 5 Appellant might have sought to obtain other documentary evidence of occasions on which customers were responsible for providing fuel. At the very least, in the absence of any other available documentary evidence, witness evidence might have been given by Mr Quinn and/or others directly involved in the running of the business, who 10 could have given a detailed first-hand account of the way the business works and the extent to which vehicles were hired on the basis that the customer was responsible for the provision of fuel, and who could have been cross-examined on that evidence. Evidence might also have been given by customers of the Appellant who provided fuel themselves when hiring vehicles. The Tribunal 15 was told that some of these customers were bodies such as education or health authorities, who presumably would have retained their own records of such matters. The letter from BCSL was far too vague to be of any assistance to the Appellant’s case.”
[99]In this paragraph the FTT explains what evidence might have convinced it that 20 there were hire contracts in which the hirer was responsible for fuelling which were not recorded in the first hire book and which would have given rise to smaller individual receipts. We do not regard this paragraph as drawing an adverse inference from the Appellant’s failure to adduce such evidence. It is simply explaining what was missing. 25100. The FTT saw and heard Mr Corr. It was able to appreciate the degree of his knowledge of the business. It did not hear, and could not test the accuracy or completeness of the statements others had made to Mr Corr. It was entitled to give little weight to the reported information from sources it could not test. The letter from BCSL 30101. The FTT referred to a letter from a Mr Quinn of BCSL which had said that BCSL had used the Appellant’s coaches on an ad hoc basis and had fuelled the vehicles. The FTT noted that Mr Quinn was also involved in the Appellant’s business, and concluded [72] that the letter was too vague to be of any assistance.[102]Evidence from Mr Quinn could have filled the gaps to which the FTT referred 35 in [72]. If the FTT had ignored Mr Quinn’s letter completely, that would have been perverse. But it did not: it said that it was too vague. To be of any use in furthering the Appellant’s case it would have had to have been shown that the coaches hired by BCSL were not recorded in the hire book. There was no indication that that was the case. It was not perverse not to regard this letter as supporting Mr Corr’s assumptions. 40 The Assessment was based on (second-hand) information provided by Mr Corr 21 We do not regard this point as telling. The FTT had to decide whether the Appellant had shown that the assessment should be different. In rejecting the evidence provided by Mr Corr the FTT was not thereby accepting the other evidence he had given to HMRC The difference, 2% was within margins 5 of estimation and error It seems to us that this would be a relevant consideration if the tribunal had decided to accept Mr Corr’s evidence that there were some hirer-fuelling hires which should have been taken into account and were not. At that stage it would have wanted to consider how accurately it could estimate the lost fuel. But the fact that HMRC’s 10 estimate of the lost fuel represented only 2% of the Appellant’s fuel consumption did not seem to us to be relevant to the question of whether there were additional hirerfuelling hires which had not been taken into account. Accordingly, it does not seem to us that failing to give weight to this factor made the tribunal’s decision unreasonable. The changes in HMRC’s calculations 15 We do not regard this as relevant: the FTT had to decide whether the Appellant had shown that the assessment was wrong or should have been different. The fact that the assessment was refined as further information was taken into account does not indicate that the assessment was wrong, even though it might indicate that further information might improve it. But there was no such further evidence. 20 103. Taking all these challenges together we ask: Was the FTT’s conclusion one which no reasonable tribunal could have reached?[104]In Rahman v C&E Comms [1998] STC 826 at 840 Carnwath LJ emphasised that in VAT appeals “the principal concern of the tribunal should be to ensure that the amount of the assessment is fair, taking into account not only the Commissioners’ 25 judgment but any other matters that are raised before them by the Appellant.”[105]To our minds that is equally applicable in an excise duty appeal. It means that if the Appellant raises an issue which the tribunal accepts is confirmed by the evidence, the tribunal must consider what effect that would have on the amount of the assessment. It is not required that the Appellant’s evidence should indicate precisely 30 what change is required. If the tribunal had accepted that Mr Corr’s evidence showed that there had been some omissions from HMRC’s calculations of hirer-fuelling hires, it would have been incumbent upon it to come to a decision as to how the assessment should be adjusted. The FTT however accepted none of Mr Corr’s evidence because it was unable to attach sufficient weight to his evidence of the assumptions behind it. It 35 was entitled to do so and so was not required to determine a different amount for the assessment. The FTT should have adopted an inquisitorial role[106]Mr McNamee says that the FTT acted unfairly in not adopting a more inquisitorial role. Whilst at [73] the FTT described the evidence which could have 40 been tendered, it made no such suggestions during the course of the hearing. He says 22 that the FTT should have asked what other evidence there was for the assumptions Mr Corr made in preparing his calculations.[107]Rule 2(1) of the FTT’s rules provides that dealing with cases justly and fairly includes dealing with cases in ways which are proportionate to the complexity of the issues and the resources of the parties, avoiding unnecessary 5 formality, seeking flexibility and ensuring that a party is able to participate in the proceedings.[108]Generally, a tribunal’s expression of its impressions on what it has (and has not) heard during a hearing may assist the participation of the parties in the hearing, and a tribunal must guard against taking too formal a view of its role. We accept that the 10 FTT could have alluded to the deficiencies in the Appellant’s evidence during the course of the hearing and invited the Appellant to adduce further evidence to fill the gaps. That would have been a desirable course of action if the Appellant had been naïve and unrepresented. But where an Appellant has competent professional representation the tribunal may fairly assume that an informed view has been taken as 15 to what evidence should be tendered and consider that there may be good reasons for not tendering other evidence.[109]Further this is an appeal made on the basis that the FTT’s decision erred in law, not an application to set aside the decision on the grounds that the hearing was not just or fair. We cannot say that the lack of contemporaneous comment by the tribunal 20 meant that its conclusions were perverse.[110]We can therefore find no error of law on the part of the FTT in relation to Ground 2 which would justify us interfering with its decision. Disposition[111]We dismiss the appeal. 25 TIMOTHY HERRINGTON CHARLES HELLIER Judges of the Upper Tribunal 30 Release Date: 5 July 2017