M J Hickey Plant Hire and Contracts Ltd v The Commissioners for HM Revenue and Customs: [2017] UKUT 0308 (TCC) [2017] UKUT 0308 (TCC)

UPPER TRIBUNAL

TAX AND CHANCERY CHAMBER

[2017] UKUT 0308 (TCC)

Case No UT/2016/0065

Between

M J Hickey Plant Hire and Contracts LtdAppellant

The Commissioners for HM Revenue and CustomsRespondent

Before

MR JUSTICE MANN

JUDGE JULIAN GHOSH

Miss Elizabeth Wilson (instructed by the General Counsel and Solicitor to HM Revenue and Customs) for Respondents

Date 31 July 2017

Category: Tax

[1]This is an appeal from a Decision in the First Tier Tribunal ( “FTT”, (

[2016] UKFTT 61

(TC) - Judge Poon and and Mr Peter Sheppard FCIS) in relation to a penalty imposed on the appellant company in respect of inaccuracies in its VAT returns from 1st December 2009 to 31st August 2013. The point in issue 10 is a short question of construction in relation to the statutory provisions governing penalties. The taxpayer company appeals the method of calculation of a penalty imposed on it, and therefore also the amount. Its appeal to the FTT failed and it further appeals to this Tribunal. 15 2. Mr Richard Stubbs appeared for the appellant taxpayer; Miss Elizabeth Wilson appeared for the Respondents (“HMRC”: she did not appear at the FTT hearing). The facts 20 3. These fall within a small compass. The taxpayer company used a computer to keep its records and prepare figures for its quarterly VAT returns. It decided to adopt what was said to be the “default” setting of the computer in relation to the dates for returns which in fact stopped the period one day short of the required period, and put that day at the beginning of the next period. The 25 result was that tax which should have been declared for that last day was declared as part of the return for the next period. This shifting went on as a consistent practice for the period of the returns identified above. The reason was said, and found, to be to improve the company’s cashflow - the VAT omitted for the last day of the proper quarter usually exceeded the input VAT 30 for that day. There were 15 such returns. When compared with what should have been declared, 9 returns resulted in under-declarations, and 6 in overdeclarations. When HMRC found out what had happened it required the then current period to be done properly, and raised a separate assessment for the last day of the previous period (31st August 2013) which would otherwise not 35 have fallen within any return.

[4]HMRC’s attention then turned to the question of a penalty. Under the provisions which will be identified in the next section of this Decision it claimed to take, as a base for the penalty, the tax that was under-declared for each period for which there was an under-declaration, and then aggregated those under-payments. This was done on a quarter by quarter 5 basis, ignoring the fact that the tax under-declared in one period was declared in a return (and in substance accounted for and paid) in the next period. That aggregate (£426,246) was then subjected to various discounts required by the statutory provisions to produce a final penalty figure of £149,186. That technique, and 10 therefore the figure, was disputed by the taxpayer. It said it should be penalised on a different basis reflecting the fact that the relevant tax was delayed but not avoided. On that basis, and applying relevant discounts, the final penalty figure would be £1,865. 15 5. There was no dispute about the underlying figures. The dispute was about the basis on which penalties should be charged, and therefore whose penalty figure was right. The relevant statutory provisions 20

[6]The statutory provisions governing penalties appear in Schedule 24 to the Finance Act 2007. They apply to a range of taxes, and not just VAT. We set out here the parts that are relevant to the appeal as it was presented to us. We do not set out paragraphs that have been rendered irrelevant as a result of 25 HMRC declining to support the reasoning of the FTT.

[7]Paragraph 1 reads:

“‘Error in taxpayer’s document 30 1 (1) A penalty is payable by a person (P) where – (a) P gives HMRC a document of a kind listed in the Table below, and (b) Conditions 1 and 2 are satisfied. (2) Condition 1 is that the document contains an inaccuracy which amounts to, or leads to – (a) an understatement of a liability to tax, or (b) a false or inflated statement of a loss, or (c) a false or inflated claim to 5 repayment of tax. (3) Condition 2 is that the inaccuracy was careless (within the meaning of paragraph 3) or deliberate on P’s part. (4) Where a document contains more than one inaccuracy, a 10 penalty is payable for each inaccuracy.”

[8]The table contains a long list of documents which we do not need to set out here. It is sufficient to say that nearly all are in the nature of returns with a few being other disclosures. They include VAT returns. 15

[9]The assessment depends on the degree of culpability of the taxpayer, and that is described in paragraph 3:

“Degrees of culpability 20 3(1) For the purposes of a penalty under paragraph 1, inaccuracy in a document given by P to HMRC is – (a) “careless” if the inaccuracy is due to failure by P to take reasonable care, (b) “deliberate but not concealed” if the inaccuracy is deliberate 25 on P’s part but P does not make arrangements to conceal it, and (c) “deliberate and concealed” if the inaccuracy is deliberate on P’s part but P makes arrangements to conceal it (for example, by submitting false evidence in support of an inaccurate figure). 30 (2) An inaccuracy in a document given by P to HMRC, which was neither careless nor deliberate on P’s part when the document was given, is to be treated as careless if P – (a) discovered the inaccuracy at some later time, and (b) did not take reasonable steps to inform HMRC.”

[10]The degree of culpability in this case was assessed at “deliberate but not concealed”, and the taxpayer did 5 not contest that.

[11]Paragraph 4 contains the penalties appropriate to the type of conduct involved:

“Standard amount 10 4 (1) This paragraph sets out the penalty payable under paragraph 1. (2) If the inaccuracy is in category 1, the penalty is – (a) for careless action, 30% of the potential lost revenue, (b) for deliberate but not concealed action, 70% of the potential 15 lost revenue, and (c) for deliberate and concealed action, 100% of the potential lost revenue …”

. Other mitigations can be applied as well; they are not material to this appeal. 20

[12]Paragraphs 5 to 8 set out the manner in which penalties are to be assessed. The important ones for present purposes are paragraphs 5 and 8, but we also set out the intervening paragraphs: 25 “Potential lost revenue: normal rule 5(1) “The potential lost revenue” in respect of an inaccuracy in a document [(including an inaccuracy attributable to a supply of false information or withholding of information)] or a failure to notify an under-assessment is the additional amount due or payable in respect of tax as a result of correcting the inaccuracy or assessment. … Potential lost revenue: multiple errors 6 (1) Where P is liable to a penalty 5 under paragraph 1 in respect of more than one inaccuracy, and the calculation of potential lost revenue under paragraph 5 in respect of each inaccuracy depends on the order in which they are corrected – (a) careless inaccuracies shall be taken to be corrected before 10 deliberate inaccuracies, and (b) deliberate but not concealed inaccuracies shall be taken to be corrected before deliberate and concealed inaccuracies. … Potential lost revenue: losses 15 7 (1) Where an inaccuracy has the result that a loss is wrongly recorded for purposes of direct tax and the loss has been wholly used to reduce the amount due or payable in respect of tax, the potential lost revenue is calculated in accordance with paragraph 5. 20

(2) Where an inaccuracy has the result that a loss is wrongly recorded for purposes of direct tax and the loss has not been wholly used to reduce the amount due or payable in respect of tax, the potential lost revenue is – (a) the potential lost revenue calculated in accordance with 25 paragraph 5 in respect of any part of the loss that has been used to reduce the amount due or payable in respect of tax , plus (b) 10% of any part that has not. … (5) The potential lost revenue in respect of a loss is nil where, 30 because of the nature of the loss or P’s circumstances, there is no reasonable prospect of the loss being used to support a claim to reduce a tax liability (of any person). … Potential lost revenue: delayed tax 8 (1) Where an inaccuracy resulted in an amount of tax being declared later than it should have been (“the delayed tax”), the potential lost revenue is – (a) 5% of the delayed tax for each year 5 of the delay, or (b) a percentage of the delayed tax, for each separate period of delay of less than a year, equating to 5% per year. (2) This paragraph does not apply to a case to which paragraph 7 applies.” 10

[13]In this case the relevant “documents” were the VAT returns for the 9 VAT quarters for which VAT was under-declared (so there was “an understatement of a liability to tax” within paragraph 1(2)(a), which was the relevant 15 “inaccuracy”, satisfying “Condition 1 In paragraph 1(2)). The inaccuracy was “deliberate but not concealed”, as we have observed above, within paragraph 3(1)(b), satisfying “Condition 2” in paragraph 1(3).

[14]The dispute in this appeal is as to whether the penalty should be assessed 20 under paragraph 5 (standard penalty) or under paragraph 8 (delayed tax). The FTT’s decision 25 15. Before the FTT HMRC maintained that its application of paragraph 5, so as to reach the large sums referred to above, was correct, whereas the taxpayer company claimed that this was a delayed tax situation and that paragraph 8 applied. The FTT reached the conclusion that HMRC was correct to assess the penalty as a standard penalty. It did so by analysing very elaborately a 30 number of other provisions in other legislation and coming to the conclusion that paragraph 8 (delayed tax) could not, by reference to those other provisions, as a matter of construction, apply to a deliberate (as opposed to careless) inaccuracy, and since the taxpayer in this case was guilty of such an inaccuracy it could not have the benefit of the lower level of penalty which the delayed tax provision attracted. Therefore paragraph 5 was correctly applied, and the taxpayer’s appeal failed.

[16]This analysis and line of reasoning was not one which had been presented by HMRC, and HMRC does not support it in this appeal. We 5 think, with respect, that HMRC is correct in that approach. We do not think that the FTT’s line of reasoning is sustainable. Instead of the arguments advanced below, HMRC took another, probably simpler, point. The taxpayer before us continued to advance the arguments advanced below, or perhaps a version of them. 10 The arguments on this appeal - an outline

[17]The arguments on this appeal can, we think, be shortly stated. 1518. The taxpayer argued that paragraph 8 applied according to its terms. The FTT found as a fact that, quarter by quarter, each inaccuracy in a previous quarter would be corrected by the inclusion of the last day of the one as the first day of the next. The inaccuracy was one and the same for the end and beginning of each period - it was not right to treat them as two separate inaccuracies on 20 each occasion. To distinguish between the two faulty aspects of each adjacent return was to create a false distinction. There was the same inaccuracy on each occasion that the “last day” was moved from one return to the next because it was produced every time by the same mechanism. 25

19. Miss Wilson’s argument for HMRC involved applying paragraph 5 in accordance with its terms and stopping there. There was an inaccuracy in each return when one day was erroneously left out of it. Where that resulted in an under-statement of tax due that was an “understatement of a liability to tax” within a paragraph 1 document, then one stopped there and calculated that 30 amount on each occasion when it happened. The “inaccuracy” at the beginning of the next period was a different inaccuracy, and did not affect the calculation made in respect of the preceding period. Taking the two inaccuracies together and treating them as one would involve taking the intention of the taxpayer into account, and there was no justification for doing 35 that at the initial assessment stage. That only came in at the later calculation stage under paragraph 4(2) when one looked at the quality of the error. She pointed to various examples given in Revenue practice statements which she said supported her case. The resolution of the arguments

[20]We consider that the arguments and case of the taxpayer are correct, for the 5 following reasons.

[21]Taken by itself, and applied literally, the wording of paragraph 5 probably drives one to the conclusion and technique contended for by the Revenue. Paragraph 1 requires (or justifies) a penalty where there is an “inaccuracy” in 10 a “document” (ie in this case a return) which gives rise to an understatement. Each return contains an inaccuracy in that it stops a day short, and that inaccuracy leads to an understatement of tax when compared with what ought to have happened (at least in 9 of the 15 returns in this case). Each of those inaccuracies would generate its own penalty calculation, though obviously it 15 produces a mathematically identical result if one aggregates those sums and works on the aggregate. Thus one arrives at the result contended for by HMRC.

[22]However, the question is not whether an unrestrained paragraph 5 can apply. 20 The question is whether paragraph 8 applies. The structure of that part of the Schedule is that paragraph 5 contains a “normal” rule, and the following 3 paragraphs provide elaborations or qualifications. Thus paragraph 8 is an alternate to paragraph 5 where it applies. The question is therefore whether it applies. 25

[23]HMRC’s case depends on applying both paragraph 5 and paragraph 8 on a document-by-document (return-by-return) basis, scrutinising each return without regard to any other return. But such an approach cannot work in relation to the sort of inaccuracy described in paragraph 8. Tax is only 30 declared later than it should have been if it does not appear in one return but does appear in a later one. So two returns are required to achieve that. The “inaccuracy” in paragraph 8 must therefore encompass two returns – in effect two inaccuracies. The first (earlier) contains an inaccuracy which wrongly results in the tax not being declared in it. The second (later) contains an 35 inaccuracy which results in the tax being declared there. The second must be inaccurate because the tax ought not to have appeared there. The paragraph therefore clearly anticipates a combined view of “inaccuracy” across two returns. It does not allow one to stop at the first inaccuracy in the manner suggested by HMRC. That first inaccuracy by itself can never result in tax being declared later. It takes a second return-type document, with an inaccuracy to do that.

[24]It also seems to us that there is a further requirement for paragraph 8 to operate. In relation to under-declared tax (here VAT), paragraph 8 requires that “the inaccuracy [that is, the under-declaration of 5 VAT in the earlier return] resulted in (emphasis added) an amount of tax being paid later than it should have been [that is, declared in a subsequent VAT return]...” An inaccuracy in a return cannot, of course, of itself “result” in the declaration of that under-declared tax in a later return. That subsequent declaration of 10 initially under-declared tax arises by reason of the relevant taxpayer’s conduct. Paragraph 8 thus requires a relevant causal connection between a taxpayer under-declaring tax in one return and that taxpayer declaring the amount of under-declared tax in a subsequent return. In other words, the taxpayer’s conduct in making an under-declaration of tax in one return must be causally 15 connected to the taxpayer declaring that amount of under-declared tax in a subsequent return. It is only then that an “inaccuracy”, being an underdeclaration of tax, can “result” in that under-declared tax being declared “later than it should have been [in a subsequent return]…” Miss Wilson referred to an example relating to trading stock, which we discuss at paragraph 25 below, 20 which illustrates how paragraph 8 applies.

[25]The degree of causation is demonstrated by comparing the facts of this case, and Miss Wilson’s accepted example, on the one hand, and the case of Miah v HMRC [2016] UK FTT 644 (TC) on the other. Miss Wilson conceded that paragraph 8 applies where, in relation to income tax, a trader under-declared 25 the value of closing stock in one return (depressing profits for that year). The trader, in Miss Wilson’s example, used the same depressed value as the opening stock figure for the next period (providing a lower base for the profit calculation, increasing profits in that second return). That, she conceded, would fall within paragraph 8. We agree. That example demonstrates first 30 how it is that two return-type documents have to be looked at, not just one. The first return (under-declaring stock) does not, by itself, result in late payment of tax. It takes the taxpayer’s conduct in completing both returns together (using the same incorrect low value for the closing stock in the first year, as for opening stock for the following year) to do that. The example also 35 demonstrates why they are linked. They are linked by the conduct of the taxpayer who puts in place a system in which the same stock figure is deliberately chosen in each of two returns. The stock figures are not included as a result of two disassociated acts. They are part of the same scheme of conduct. The same is true of the present case. The shifting of each return by 40 one day, quarter by quarter, is a system adopted as a result of the conduct of the taxpayer which creates a relevant causal link between each successive return and the preceding one. This results in the late payment of tax (where that is the effect of the figures) in each quarter, combining the effect of the two returns.

[26]This situation is to be contrasted with that in Miah v HMRC [2016] UK FTT 644 (TC). In that case a taxpayer ought to have accounted for VAT on a supply in May 2014 in the quarter to which that supply related (June 2014). However, he did not include it until a return in March 2015, when the proceeds of the supply were made available to him. 5 The FTT rejected a submission that paragraph 8 applied:

“61. On reflection it seems to us that this provision does not apply. It applies only if the inaccuracy itself "resulted" in a later 10 in an increased declaration [sic]. Its focus is on the inaccuracy rather than the conduct which gave rise to the inaccuracy. It does not apply where the delay arises solely because the taxpayer's conduct has had the effect of an increased declaration in a later period. The kind of inaccuracy with which 15 the paragraph is concerned is that for example which may arise if closing stock has been understated. That has the automatic affect it - it ‘results’ in - the opening stock being understated and the profit overstated in the subsequent period. In Mr Miah’s case the declaration of the sale in the later period was a 20 result of the decision to declare it in that period rather than the lack of recognition in the earlier period.”

[27]In that case there was nothing like a system whose substance was to shift the tax from one particular period to another. There was merely a sequence of 25 events, which eventually resulted in a declaration in a later period. When he did not declare the tax in the correct period it was not clear when it would be declared. He then declared it when he chose (when he received funds). The under-declared VAT for the earlier VAT quarter had no causal relationship to the subsequent declaration of the under-declared VAT (which was found to 30 arise from the trader having been put in funds). In those circumstances it is proper to view the situation as containing two discrete inaccuracies, where the taxpayer’s conduct in under-declaring VAT in one return had no connection to (did not “result in”) a subsequent declaration of that under-declared VAT. Incidentally, we observe that, in this case, an amount of under-declared VAT 35 for one quarter is indeed declared late if that amount is included in a later return. The fact that the later return may under-declare a different amount of VAT (arising from the turnover on the last day of the later period) does not disturb the observation that the under-declared VAT from the earlier period has indeed been included (and therefore declared) in the later return. 4028. The effect we have described seems to us to be the plain effect of the paragraph. Our view is strengthened by two factors. First, were that not the case, and if HMRC were allowed to stop at the first inaccuracy and base the penalty on the tax undeclared in that return, without any regard to any subsequent return, then it is hard to see how paragraph 8 would ever have any work to do. Second, the analysis is actually accepted by HMRC in its acceptance of the closing stock example, which we have referred to above. Miss Wilson sought to justify her conclusion, and a distinction from the present case, by saying that her stock example was one 5 where there was an inevitability of an under-declaration because the low stock value in the first return would have to lead to the same figure appearing in the second return as opening stock, because that is what the system required. But it is only “inevitable” because that is what the taxpayer has chosen to do, in completing 10 the two returns. It is the system adopted by that particular notional taxpayer for the two returns. He or she could, in theory, put an entirely different opening stock figure in the second return, though that would unlikely because he/she would immediately be caught out. There is no factual inevitability of the under-declared tax being declared in the later return because the trader had 15 used an incorrect low value for closing stock for the earlier year; there is only an inevitability because of the commercial system adopted, and followed through, by the taxpayer, of using the same incorrect low figure for closing stock in the earlier year as for opening stock for the following year. 20

29. So the question in any individual case is whether it is right, in substance, looking at all relevant facts, to view the two inaccuracies as an inaccuracy (in one return) leading to (resulting in) the late declaration of tax (in a subsequent return). That is a question of fact. In the present case the clear findings are that the taxpayer adopted a system which fell within that description. Return 25 by return, the taxpayer made one day fall out of one period and into the next. This was a consistent and systematic approach. The result was (in some cases) the late declaration of tax in relation to that day. That is, in our view, an inaccuracy which has resulted in the declaration of tax being late, within paragraph 8. It has been shifted from one quarter to the next. While it is true 30 that there are two inaccuracies (one in each of two returns) it must be the sort of “inaccuracy” which paragraph 8 contemplates.

[30]This interpretation of the paragraph seems to us to be completely in accord with, and indeed required by, what is accepted to be the purpose of the 35 paragraph (to base the penalty on the time value of money in a true delay case) and in accordance with the purposive approach to statutory interpretation set out in many authorities, including Pollen Estate Trustee Co Ltd v HMRC

[2013] 1 WLR 3785

at paragraph 24 (Lewison LJ).

[31]We address a further submission made by Miss Wilson. Miss Wilson accused 40 the appellant of “penalty shopping”, which we assume is a reference to an attempt to manipulate the penalty code so as to attract the lesser penalty in Paragraph 8 in circumstances where the taxpayer’s conduct was culpable and deserved the greater penalty imposed by paragraph 5. We do not understand this submission. 6 of the 15 VAT quarters in question resulted in an over declaration of VAT so the appellant, for these 6 VAT quarters, was not “shopping” for anything at all. And, more fundamentally, the culpability of the taxpayer is assessed and dealt with by Paragraph 5 4 (depending on whether the inaccuracy was “careless”, “deliberate” etc.), which imposes a greater penalty for greater degrees of culpability. So conduct relating to “penalty shopping” is dealt with by Paragraph 4. Conclusion 10

[32]We therefore determine that the appeal should be allowed and the penalty calculated in accordance with paragraph 8 and not paragraph 5. The figures which apply in that eventuality have been agreed, so we do not have decide anything about them. 15 33. The parties should make any applications for and about costs, and any ancillary applications, following on from the sending out of this Decision. 20 MR JUSTICE MANN JUDGE JULIAN GHOSH 25 RELEASE DATE: 31 July 2017