WM Morrison Supermarkets PLC v The Commissioners for HMRC

Under this ground the appellant argues the FTT was wrong to consider the healthiness of the product, and its marketing as such, as irrelevant to the question of whether the products were “confectionery”.

The FTT set out its reasoning, that healthiness was irrelevant as follows:

“Healthy food?

174.

Mr Simpson [for the appellant] said that the customers for the Organix Bars were health conscious, and that the Organix Bars were marketed as a healthy alternative to confectionery. Mr Watkinson said this was irrelevant. He relied on Kalron, which considered the VAT status of smoothies made from liquified fruit and vegetables. In that judgment, Warren J first set out the excepted items under the heading of “food” in Group 1 of Sch 8, followed by the overrides to the exceptions and the Notes, and then said:

“[9] It is difficult to detect any policy behind these detailed exceptions and overrides. Mr Thomas (who appears for Kalron) claims to identify a policy which is to exclude what he calls junk food…

[10] It is impossible, in my judgment, to spell out of the structure and content of Group 1 a policy such as Mr Thomas submits can be detected. There are plenty of ‘junk’ foods which do not fall within the exceptions; and there are healthy drinks which are within the exception, for instance, freshly squeezed orange juice.”

175.

I respectfully agree. I also agree with the FTT in Corte Diletto when they said at [100]:

“The healthiness or otherwise of a product has no bearing on its VAT classification. Zero-rating depends solely on whether a product has sufficient characteristics to fall within one of the Groups in schedule 8 when applying the [statutory] tests…The most sugary, cream filled, chocolate covered cake will still be zero rated. The healthiest of low sugar, low fat confectionery will still be standard rated.”

176.

I therefore agree with Mr Watkinson that whether or not a product is healthy rather than “junk” food is irrelevant when carrying out the multi-factorial test necessary to decide a product’s VAT status, and I have not taken it into account below”.

The appellant argues the FTT:

(1)

misapplied Kalron: that case dealt with the different question regarding the absence of any obvious policy behind Group 1 as a whole;

(2)

wrongly excluded healthiness as a factor because it was not determinative – that it was not determinative did not mean it was not relevant (as the FTT accepted in relation to other factors such as packaging which were accepted to be relevant even if not determinative); and

(3)

failed, in line with the case-law, to consider the relevance of how the product was marketed.

In response, HMRC highlight that the legislation, and in particular the deeming provision of Note 5, makes no reference to healthiness. Regarding Kalron, the FTT correctly interpreted that as holding that healthiness was irrelevant.

In Kalron the High Court (Warren J) considered the question of whether smoothies made from liquefied fresh fruit and vegetables fell within the term “beverage” within item 4 of the excepted items in Group 8 of Schedule 1 VATA 1994 and therefore standard-rated for VAT purposes, as HMRC argued, or, as the appellant argued, zero-rated as “food of a kind fit for human consumption” as referred to in Group 1(the same category as the appellant contends for in this case). The taxpayer lost before the VAT tribunal and appealed to the High Court. It is helpful to set out [9], which the FTT in the instant case quoted in part, in full.

“It is difficult to detect any policy behind these detailed exceptions and overrides. Mr Thomas (who appears for Kalron) claims to identify a policy which is to exclude what he calls junk food: thus ice-cream, confectionary and crisps are excluded and become standard rated. He suggests that the paradigm beverages within Excepted item 4 are branded fizzy drinks typically bought in cans or plastic bottles which can, again, be seen to be in the nature of junk drinks. On that basis, he says that healthy products such as the Product should, if there is a doubt about their status, be put the non-beverage side of the line.”

That paragraph came under the heading “A preliminary point” in a section which appeared in the part of the decision before the High Court went on to summarise the VAT tribunal decision and discuss the grounds of appeal.

In his discussion of the error alleged by the taxpayer that the tribunal was wrong to dismiss product’s ingredients and nutritional effects as irrelevant, Warren J noted (at [40]) that the tribunal did not any case say those factors were irrelevant. Referring back to the “anti-junk food policy” argument described above he also said:

“this aspect of the case takes on less significance, I think, given my rejection of Mr Thomas’[counsel for the taxpayer] submissions in relation to the policy of the inclusion and specification of the Excepted items in Group 1 for the reasons already given”

The argument was again referenced at [51] where Warren J set out:

“Mr Thomas then submits that the meaning of beverage must be construed with regard to the purpose of the statutory provision in which it is contained. I agree with that insofar as a clear purpose can be ascertained. But I reject, for the reasons already given under the heading "A preliminary point", his submission that the exception is to exclude items of little or no nutritional value”

Ms Sloane argues the FTT mis-read Kalron in considering that healthiness was irrelevant. Kalron firstly was dealing with the different issue of what was a “beverage”. Secondly, and more significantly, there was no argument about the ordinary meaning of that term from the point of view of the ordinary person on the street. The appellant was arguing that healthiness (although it was described in terms of a junk food/ non-junk food distinction) represented the underlying policy underpinning what items were respectively standard-rated and zero-rated. That was the argument which the High Court rejected. The court did not hold that healthiness was irrelevant to the “person in the street” test.

Regarding the appellant’s first point of distinction, we agree with Mr Watkinson, the fact the case was about a “beverage” did not matter because the point the taxpayer sought to make was a more wide-ranging argument concerning the standard or zero rating of food and drink items more generally. We also agree with him, that the taxpayer’s argument was not dealt with as some kind of interesting policy debate which took place in a vacuum; the argument was put forward to advance the taxpayer’s case that the product in question, being healthy, was zero-rated.

The question remains whether, the case stood for the proposition, as the FTT held it did, that healthiness is irrelevant.

In our view, looking at the way the taxpayer put its argument in Kalron, it is clear it was not simply saying healthiness was relevant; it was saying it would, in a case of doubt, be a deciding factor in putting a drink which was healthy on the non-beverage side of the line. Ms Sloane suggested the appellant in Kalron was effectively using healthiness as a “trump card”. We think their argument, being confined to cases of doubt, was slightly more nuanced and can better be described as seeking to place an interpretative gloss on the meaning of terms in Group 1 such that standard rated items were junk food/ junk drinks, and zero rated items were healthy. This was this argument that Warren J roundly rejected in his preliminary discussion. But, in agreement with Ms Sloane, we do not see the case stands for the proposition that when considering the issue of fact of whether something was a beverage, healthiness was irrelevant; rather he rejected the argument that healthiness was a determinative factor.

We therefore agree with the appellant, that Kalron, in rejecting the argument that there was an anti-junk food theme which coloured the interpretation of the VAT provisions dealing with food and drink items, did not rule out considerations of healthiness when considering whether a product fell within the ordinary meaning of the relevant item.

HMRC also relied on two FTT cases which it was submitted reached the conclusion that healthiness was an irrelevant factor. (Mr Watkinson originally advanced these to demonstrate that the FTT’s disregard of the factor in the current case, could not be regarded as perverse, given other FTTs had also considered it irrelevant – however for the reasons explained that argument is no longer relevant given our conclusion that it is not necessary for the appellant to show an additional hurdle of perversity).

In Corte Dilletto UK Limited v HMRC [2020] UKFTT 75 (TC) (mentioned in the extract from the FTT Decision above) the FTT concluded the product was confectionery. The product, “Nouri Truffles” later called “Nouri health balls”, were small balls made from dates, nuts and other natural ingredients with no added sugar. Amongst the submissions the FTT summarised (at [46]) was that “As [the products] are healthy they are fulfilling a social policy of encouraging people to cut down on sugar and should be zero-rated”. Although Ms Sloane pointed out the section quoted by the FTT in the current case (at [77]) above appeared in the section of its decision under the hearing “Does social policy have an impact?” and after the multi-factorial evaluation where the FTT considered the product was confectionery, we do not consider anything significant can be drawn from that. The ordering was consistent with the FTT considering healthiness to be entirely irrelevant to the multi-factorial evaluation: the FTT was not referred to Kalron but ruled (at [100]) “the healthiness or otherwise of product has no bearing on its VAT classification” and then went on to give the reasoning quoted in the FTT Decision extract above.

We agree with Ms Sloane the appellant was relying on healthiness as a “trump card”. Not only did the FTT in Corte Diletto reject that, but it seemed to us go further in saying healthiness had no bearing, a proposition on which the FTT in this case relied. However, we note that although the examples provided of unhealthy zero-rated items, and relatively more healthy items which were nevertheless standard rated, could explain why healthiness could not be determinative, it would not explain why healthiness could not at least be relevant in a multi-factorial evaluation. As Ms Sloane submitted, many of the factors the FTT in this case relied on (e.g. packaging) as part of its multi-factorial evaluation were considered relevant even if they were not determinative.

The other FTT case HMRC rely on was Innocent Ltd v HMRC [2010] UKFTT 516 (TC) where the issue was whether the product – pure fruit smoothies made of 50% fruit juice and 50% fruit salad – were “beverages” under the same provisions as considered in Kalron. The FTT considered itself bound by Kalron in terms of that case establishing that the legislation did not bear out an anti-junk food policy, and stated that would in any event have come to the same view ([21]).

In our view, Innocent does not assist HMRC. Having found (at [26]) there was no consistent social policy behind the exception for certain beverages, the FTT went on to consider the ordinary meaning of beverage (which suggests that the FTT viewed the social policy argument put to it as an argument that the meaning of beverage should be legally glossed). In rejecting HMRC’s argument that Kalron was binding on it for the proposition that fruit-smoothies were beverages, the FTT noted the evidence in relation to the products was not the same in particular (at [68]) the evidence on “marketing and on the nutritional effect of smoothies”. At [69] the FTT noted Warren J had agreed with the taxpayer’s counsel that the tribunal ought to consider questions which included “nutritional value”. In that respect, the case is supportive of healthiness being relevant (even if not determinative). It is certainly not inconsistent with healthiness being at least a relevant factor.

In seeking to explain why the FTT’s decision to disregard healthiness was not perverse (a hurdle which as we have discussed is not required) HMRC also argued healthiness could not be a relevant factor of any weight which would outweigh the other factors pointing towards confectionery as mentioned in Note 5. We had difficulty understanding this line of reasoning given Note 5 does not purport to be exhaustive of the criteria relevant to determining whether something is confectionery. The point of a deeming provision, such as Note 5, is that if the specified criteria are met, it saves having to agonise over whether a product is confectionery or not. However, we consider nothing of significance can be taken from the absence of a factor in the deeming provision’s list as regards the relevance of that factor. One could equally put another factor, though not specifically mentioned in the legislation, such as packaging, and say that because that is not mentioned in Note 5 it would not outweigh the specific factors listed in the deeming provision. However, that would not preclude it being relevant on the question of whether the product was “confectionery”. Moreover, as to the factor’s weight, whether a particular factor would outweigh the cumulative weight of the factors listed in Note 5 will depend on the particular facts of the case and how the matter is looked at in the round. We reject any suggestion that, as a matter of legal principle, a factor which is not mentioned in Note 5 is presumed to be of less weight than the cumulative weight of factors which are so mentioned.

Regarding the appellant’s allegation that the FTT conflated the issue of whether a factor was relevant with the factor being determinative, Mr Watkinson points out that this error is not made out because no such assumption is mentioned anywhere in the FTT Decision. While the FTT did not express its reasoning explicitly in those terms, the FTT did in essence rule out healthiness on this basis: it saw Kalron as saying healthiness was not determinative and therefore not relevant. As Kalron did, in our view, confirm that healthiness was not determinative, we do not however see this as a discrete error but part and parcel of the error the FTT made in holding that Karlron stood for healthiness being irrelevant.

HMRC further argue there is good reason not to import considerations of actual and purported healthiness into the assessment of whether something is confectionery. Healthiness, they argue, is a loose concept, the standards by which it is to be measured are unclear; it would make the multi-factorial assessment unworkable. We reject this submission. Many of the factors accepted to be relevant, for instance, whether packaging is brightly coloured are similarly not defined according to precise standards and rely on the subjective judgment of the fact-finding tribunal. Courts and tribunals have in practice made evaluations on healthiness without difficulty: in Kalron Warren J, in the course of his reasoning rejecting the appellant’s “anti-junk food policy” argument gave an example of a healthy drink (freshly squeezed orange juice) (at [10]). In Corte Diletto the tribunal (although it did not regard healthiness as relevant) was able to reach a view (see [23]) on the product’s healthiness by reference to the food “traffic light system” and information on fat and sugar content from the British Nutrition Foundation and NHS websites. Furthermore, as Ms Sloane pointed out, in Premier Foods ([2007] VAT Decision 20072) the FTT did place weight on the product marketing’s healthiness in terms of the lack of added sugar, artificial ingredients and flavours and its low fat content, and to recommended portions of fruit, features it did not consider typical of confectionery (at [29]). That reasoning was not criticised on the subsequent appeal to the High Court (we come on to cover that decision under Ground 2 and in dealing with the arguments HMRC make in its response on the applicability of Note 5).

We conclude there is no reason in principle why healthiness was not a factor to be weighed up along with all the others in the balance when considering how the ordinary person on the street would view the product.

Mr Watkinson further suggested the appellant’s reliance on the ordinary person the street test was misconceived. He emphasised the test was concerned with what an informed person on the street, in other words that the person would know, as the tribunal did, that there was no policy regarding healthiness which was relevant to the classification of the product. However, this argument is predicated on HMRC’s view on the significance of Kalron being correct which, for the reasons we have already explained, is a view which we reject. We also bear in mind Toulson LJ’s observation in Pringles about the test (at [63]) and his note of regret that it had led to “distracting argument about what knowledge should be attributed to [the] hypothetical person”. In agreement with Jacobs LJ, he considered the test “…really amounted to saying no more than that it was for the tribunal to decide what was the reasonable view on the basis of all the facts known to the tribunal; and it conveys that this is not a scientific question.” We respectfully agree. It does not seem that the case-law principles invite consideration of what knowledge of legal principles should be imputed to the informed person on the street, but even if that was what was required such person, would simply be taken to know that, while there was no rule of interpretation that healthy foods were presumed not to be standard rated, that did not mean that healthiness was irrelevant.

In relation to the appellant’s criticism that the FTT failed to apply the case-law principles, which establish that how a product is marketed is a relevant factor, we are not persuaded the FTT made such error or at least any additional error. The FTT clearly did consider the product’s marketing, it just wrongly excluded healthiness aspects from that consideration in accordance with its self-direction that healthiness was irrelevant.

HMRC accepts the FTT did direct itself to exclude the factor of whether or not the product was healthy or junk food. However, Mr Watkinson took us to various excepts in the FTT’s summary of evidence and in its multi-factorial evaluation for both the Organix and Nakd bar products, which in his submission, showed the FTT did not actually then exclude the products’ marketing as healthy, or its perceived healthiness from consideration.

Regarding the Organix product, at [121], the FTT recorded evidence on behalf of appellant that these “… were held out for sale [by Morrison’s] as healthy snacks”, which it then referred back to in its multi-factorial evaluation (at [177](5)(e)). When describing the packaging it mentioned what was said about the products being organic and a “no junk promise” ([125][126]).

Similarly in relation to the Nakd bars, the FTT mentioned the words on the packaging “100% natural ingredients” ([194]), the positioning by the “healthy biscuits and cereals section” and then the “healthy snacks” section ([195)), the description on Morrisons’ website mentioning “…healthier cereal bars” “…healthier options”, ([196]) and further references to the product being healthy in Nakd’s website marketing ([197]). In its multi-factorial assessment at [206] it mentioned customers referring to the product as “perfect for those moments when you want something sweet and healthy” (2), it acknowledged Nakd bars were positioned as “being more healthy than traditional products” (8) and the products’ eventual placement with “healthy biscuits and cereals” ([13]). Mr Watkinson accordingly submits the FTT did take account of healthiness. That then meant, so the argument ran, that the appellant’s challenge could only concern the lack of weight given to the factor, challenges in relation to which, it was agreed, would not constitute and error of law.

Regarding perceived healthiness Mr Watkinson took us to the passages in the FTT decision ([129] – [131]) where it accepted his submissions regarding the inadequacies of the evidence the appellant put forward (the “basket association analysis” advanced by a manger in the “customer insights” team) in showing how customers viewed the products. Mr Watkinson suggested the FTT would not have analysed the insufficiency of that evidence in this way if it regarded perceived healthiness as wholly irrelevant. In the end the FTT considered there was little that could be drawn from that evidence. However, the fact that findings were not thereby made on the issue, which was a failure on the part of the appellant, could not be described as the FTT failing to take something into account.

We reject HMRC’s submission that the FTT did, despite its self-direction to the contrary, take into account healthiness and perceived healthiness in the way suggested. The first point is that the FTT’s clear self-direction was that healthiness was irrelevant as a factor in its overall assessment. It ought, in our view to be presumed that the FTT will act in accordance with its own direction. We consider it did. We agree with Ms Sloane that the fact the word healthy or aspects of that concept appeared in the extracts HMRC refer to did not mean healthiness was (contrary to the FTT’s self-direction) being considered.

The fact the FTT mentioned aspects of healthiness in its summary of the evidence, and its analysis of such evidence, does not mean it then regarded those points as relevant. It is also important to recognise, as Ms Sloane pointed out, the context in which the word “healthy” did get mentioned in the FTT’s multi-factorial assessment. The reference regarding Organix bars being held out as healthy snacks at [177(5)(e)] was advanced to justify the FTT’s finding there that the bars were held out as snacks– the snack’s healthiness was not considered. For the Nakd bars, the excerpt at 206(2) (“…something sweet but healthy”) was referenced in the context of a discussion under the heading “sweet to taste”, so the relevance lay in the reference to “sweet”. While at (7) and (8) the FTT accepted the bars were sold as healthy and positioned as being more healthy than traditional products, the FTT considered this did not detract from them being held out as treats or held out as filling the same role as traditional confectionery. Regarding placement, the point the FTT took from the evidence was that the products were subsequently placed with “healthy biscuits and cereals”, was that biscuits were with the normal meaning of confectionery.

In conclusion, we are satisfied Ground 1 discloses an error of law in the FTT Decision.

HMRC also rely on an argument that as there is “no ideal concept, conformity of every aspect of which is necessary, before a product can be called a certain description” (Customs and Excise Commissioners v Ferrero UK Ltd [1997] BVC 408 per Lord Woolf MR and Hutchison LJ) and that there is, therefore, even less scope to argue that a tribunal has ignored a relevant factor as part of its multi-factorial assessment. We do not consider this point, which in essence is that one should not expect there to be an exhaustive check-list of factors, as assisting in the circumstances of this case. This is not a situation where, after the FTT performed its evaluation the appellant sought to advance another factor said to be relevant, where the factors the FTT had already considered were sufficient. Where the FTT has specifically and incorrectly ruled a factor to be irrelevant that will not prevent the omission from constituting an error of law. The argument is really one about the materiality of the error, an issue we consider later.

Under this ground, the appellant argues the FTT erred in law in failing to consider that the absence of ingredients associated with traditional confectionery (cane sugar, butter or flour) was a relevant factor. The appellant raised this argument before the FTT in relation to both the Organix and Nakd bars.

In relation to the Organix bars the FTT explained (at [177(8)

“As noted above, the products considered in Premier were “fruit bars” and the sugar content was derived from the dried fruit, but they were nevertheless confectionery. The absence of cane sugar, flour and/or butter is therefore not a factor pointing to the Organix Bars falling outside the meaning of confectionery”

In relation to the Nakd bars, the FTT dismissed the relevance of the absence of such ingredients for the same reasons ([206(12)]).

The FTT therefore relied on Premier Foods to justify the irrelevance of the point. We can deal with this point relatively briefly. The short point, which Mr Watkinson accepted, is that the litigation in Premier Foods did not decide the fruit bar products there were confectionery.

We will return to the case below in dealing with the Note 5 issue, but for present purposes it is sufficient to note the following. The High Court’s decision was an appeal by HMRC against the VAT Tribunal’s decision that the fruit bar product was not confectionery. Part of the Tribunal’s reasoning concerned the lack of added sugar, and that the production process did not involve cooking. The High Court at [17] agreed with HMRC’s case that neither cooking nor added sugar were necessary for something to be confectionery. It concluded that the Tribunal had made errors of law and remitted the matter to a differently constituted Tribunal for a fresh assessment of whether the fruit bars were or were not confectionery (but there is no report of what that the new Tribunal ultimately decided). The FTT was therefore wrong to rely on Premier Foods as a basis for excluding the relevance of the point.

In line with their case that there was an additional perversity hurdle to surmount HMRC put their case in terms of the FTT’s view the ingredients were irrelevant as not perverse. Given our decision above that there is no such hurdle, we only need to be satisfied that consideration of the absence of such ingredients was at least relevant.

Mr Watkinson pointed to a number of instances of items accepted to confectionery e.g. sweets, which lacked butter and flour. Also, he suggested, it would mean that if any accepted item of confectionery such a sweet had the cane sugar in it replaced by artificial sweetener then the result would be different which was obviously wrong. Mr Watkinson again sought to test the relevance by positing whether when weighed against the named factors in Note 5 it would lead to a different result. We reject that for similar reasons to those discussed above. HMRC’s argument also assumes the nature of the ingredients needs to be determinative to be relevant. As with a number of factors the FTT did take into account (such as packaging or marketing), they were relevant even if they were not determinative.

We therefore find there was an error of law in the FTT rejecting the relevance of such ingredients. In agreeing with the appellant, that the FTT erred in law, it is important not to overstate the relevance of such traditional ingredients and to elevate their presence or absence into an essential characteristic. A consideration of whether something is confectionery will inevitably involve comparison with products which are present in items commonly accepted to be confectionery. There will no doubt be examples of confectionery which do not contain such ingredients but which are nevertheless confectionery. But that does not mean consideration of the ingredients, and the absence of traditional ones, will not add to the overall picture of the product’s classification.

We will consider the materiality of the errors we have found under Grounds 1 and 2 once we have dealt with HMRC’s argument in response on Note 5. That is because if they are correct in their case, it would mean the products would be deemed to be “confectionery” and the appellant’s appeal would fail in any event.

Note 5 is set out above at [5]. It deems products with certain attributes: (1) prepared food (2) normally eaten with the fingers (3) sweetened to fall within the “confectionery” exception under Item 2 of Group 1 Schedule 8 VATA 1994. The appellant accepts both products meet (1) and (2), the issue relates to (3), viz the meaning of the word “sweetened”. Does it cover products which are inherently sweet (as HMRC argue), or does it require that the sweetness is added (as the appellant submits)?

The FTT rejected HMRC’s case (at [94]–[106]) that Premier Foods stood as authority in support of their interpretation. In their response, HMRC submit the FTT was wrong to do so and make further arguments not advanced before the FTT in support of their position.

As mentioned above, in Premier Foods the product was a fruit bar whose sugar content was derived from dried fruit. At first instance the VAT Tribunal decided the bars were not confectionery. In interpreting the term “confectionery” the Tribunal, taking account of a dictum of Lawton J in a decision, Popcorn House Ltd [1986] 3 All ER 782, suggested confectionery was made with a cooking process involving heating and involved the ingredients being made sweeter than their natural state ([26] Premier Foods VAT Tribunal). The tribunal noted the primary ingredients, the fruits, were intrinsically sweet and that the heating process (pasteurisation of certain ingredients prior to their addition) was only in relation to a small proportion of the whole.

On their appeal to the High Court (Sir Andrew Morritt C), the Commissioners argued the Tribunal was wrong to apply Lawton J’s dictum given it related to materially different legislation (Group 34 of the Purchase Tax Act 1963). The High Court agreed, (noting the Tribunal had also misread Lawton J’s dictum as suggesting recourse was to be had to the ordinary meaning of the term). The Commissioners also argued (at [9]) the Tribunal “were wrong to conclude that confectionery, for the purposes of excepted item 2 and note (5), must include additional sweetening matter” (HMRC emphasis) and that the product must have been cooked in some way.

The Chancellor rejected the submissions at [17]:

“… the Tribunal clearly directed themselves that for an item to be classified as confectionery for the purposes of excepted Item 2 and Note 5, its production must have involved (a) a process which can be recognised as cooking and (b) the addition to the primary ingredient of an extra element as sweetness. In my judgment, neither of those elements is a necessary condition for a product to be classified as confectionery…”(HMRC emphasis)

On the basis of the three errors, the appeal was allowed and the matter remitted to a differently constituted tribunal to determine the classification of the fruit bars afresh ([19])

HMRC rely on the reference, in the Commissioners’ submission, to Note 5 as well as item 2 (at [9]) and repeated references to both “excepted Item 2 and Note 5” at [11], [16] and [17].

Mr Watkinson argues there would have been no need for the Chancellor to refer to Note 5 at all if all the High Court had been doing was addressing whether a product needed to be sweetened and cooked to be confectionery within the meaning of Item 2 (without recourse to the deeming provision in Note 5).

We reject that submission. Although HMRC rely on the Chancellor’s conclusion at [17] that passage actually confirms that the Chancellor’s focus was on the meaning of the word “confectionery”, not “sweetened”. This is reinforced by the Chancellor’s elaboration of his reasoning later in the same paragraph where he accepted:

“..in its ordinary usage, confectionery is limited to products which can be described as sweet but I cannot see why such sweetness may not be inherent in the principal ingredient in its natural state but must be added by some further sweetener with which it is mixed or compounded…it appears that…the Tribunal erred in law in considering those two elements [cooking and addition of sweetener] were essential to the categorisation of these fruit bars as confectionery”

It is clear all the references to Note 5 as well as Item 2 both in recording HMRC’s submission before the High Court (at [9]), and in the following reasoning were included because Note 5 also includes the term “confectionery”. The fact it was this term (not “sweetened”) whose interpretation was in issue is all the more evident, as Ms Sloane points out, from the scope of the Tribunal’s decision under appeal in that case where the Commissioners were arguing the normal meaning of the word “confectionery” (“in excepted item 2 without regard to Note (5)”) covered “prepared food which [was] rich in sugar although not sweetened”.

If the High Court were, in addition, intending to set forth an interpretation of the term “sweetened” in Note 5 we would have expected, as Ms Sloane argued, that it would address that issue explicitly. This is especially so, given the interpretation, which HMRC say the High Court judgment gave, does not accord with the normal meaning of “sweetened”. The use of the past participle, when the legislation could simply have referred to “sweet” if that was what was intended, indicates Parliament had in mind products with added, rather than inherent sweetness.

We accordingly consider the FTT correct to reject HMRC’s argument for the reasons it did. We were also not persuaded by HMRC’s further arguments.

If we were wrong in our reading of Premier Foods and the decision is to be read as standing for the proposition for which HMRC argue , then we would, in any case, be convinced it is wrong for the reasons already outlined above (the ordinary meaning of the term “sweetened”) such that we would not follow the decision (see discussion at [94] in Gilchrist v HMRC [2014] UKUT 0169(TCC) which considered the precedential effect of High Court decisions on the UT).

In so far as other tribunals (H5 Ltd (t/a High Five) v HMRC [2008] UKVAT V20821 at [24] – [26], Corte Diletto at [69] – [71]), interpreted Premier Foods as holding that the requirement “sweetened” in Note 5 was met if the product could be described as “sweet” then, in agreement with the FTT, we consider that those tribunals were wrong to so hold.

HMRC argue both errors alleged under the Grounds 1 and 2, even if made out, are immaterial. Under Ground 1 they submit there was no or insufficient evidence of objective healthiness and argue that in any case the omitted factor would not outweigh all the cumulative factors identified by the FTT which justified classifying them as “confectionery”.

We reject HMRC’s argument that there is no evidence regarding healthiness. As Ms Sloane explained, there was sufficient evidence which emerged from the evidence regarding the nature of the ingredients. As confirmed in the evidence on ingredients, these were wholly natural and did not contain any added sugars or syrups. As she pointed out, it was not necessary to present scientific evidence to a tribunal for it to reach a view on a product’s healthiness. We have already mentioned above how the High Court in Kalron and the FTT in Premier Foods (see [96]) were able to take a view on healthiness without such evidence. We thus agree with Ms Sloane that there was sufficient evidence before the FTT for it to do so too.

For both errors HMRC argues neither outweighs the cumulative weight of all the other factors and submits the result would therefore be no different. We have already explained why a “would have been different” test is wrong. The test is whether the decision “might have” been different (in line with Degorce). We assume Mr Watkinson would say that even that “might have been different” test is not satisfied because of the cumulative weight of the other factors. However, we agree there are difficulties in that assumption for the reasons highlighted by Ms Sloane.

First, it must be recognised that the multi-factorial evaluation cannot be reduced to an exercise of counting up the numbers of factors which point in a particular direction. Rather, the exercise is qualitative and it is possible that because of the weight ascribed to a particular factor in forming an overall impression it outweighs a number of other factors which point in the other direction. We did not understand Mr Watkinson to disagree with that in principle. Healthiness could be of insignificant weight or could be of great significance, say for instance the tribunal were to accept the significance of the appellant’s argument that because the bars were a healthy alternative to confectionery they were not thus confectionery. In relation to a comparison with traditional ingredients, Ms Sloane illustrated this point with a hypothetical vegetable crisp product. By reference to the various criteria the FTT used there might well be a great number of factors which pointed to it being confectionery, for instance, a high sugar content because of naturally occurring sugars, its sweetness, that it was subjected to a process, that it was normally eaten with fingers, that it was consumed as a snack, that its packaging was bright and colourful. But if one left out the factor that its ingredients consisted mainly of vegetables, and that it did not contain ingredients traditionally associated with confectionery and went on to find that because of the other factors the product was therefore confectionery that would as Ms Sloane submitted, “untether” the term confectionery from its ordinary meaning.

Second, we agree healthiness is a factor which may have a pervasive effect in the sense that it colours the impression of other factors. For instance, as Ms Sloane pointed out the FTT’s analysis of the packaging indicated that while the Nakd bars were sold as healthy food they were also positioned as treats ([206(7)]) did not mention all the various wording which highlighted the healthiness of the product to balance against the wording the FTT did refer to which was suggestive of the products being a treat. The FTT accordingly did not, grapple with the point that products, for example a raw date and nut bar, which are healthy, are not typically regarded as treats.

We can also see how, as Ms Sloane submitted, taking account of the nature of the ingredients and their healthiness or perceived healthiness (raw cold-pressed fruit and nuts) might colour the impression one gains from tasting the product and in particular whether its taste is like that of traditional confectionery ( for instance sweetness).

We are satisfied that the errors (of failing to take account of healthiness and/or the nature of the ingredients by comparison with traditional confectionery) might have made a different to the decision if they had been taken into account.

While it is relevant, to consider whether the decision would be sustainable on the basis of the unchallenged findings made, we do not consider a decision the products were confectionery would be possible, on the basis of the findings which can be isolated, given the qualitative nature of the findings and the pervasive effect of the impugned factors on the other findings. The remaining findings on sugar content, that the ingredients were subjected to a process, were normally eaten with fingers, were held out as snacks, consumed between meals, and their names (placement being at best a neutral factor) would not, we consider, be sufficient by themselves to sustain a holding the products were confectionery. It would be necessary to reach conclusions on the impugned factors of taste, whether they were held out as treats, whether they were held out as filling same role as traditional confectionery, their packaging and ingredients too. Those factors would, in line with our discussion on pervasiveness above, also include considerations of the healthiness of the products and that they lacked the ingredients associated with traditional confectionery.

We therefore consider the decision should be set aside. It is worth emphasising that while the materiality of the errors is sufficient to set aside the decision on the basis we are satisfied that but for the errors the decision might have been different, that of course leaves open the possibility that a tribunal, whether us or the FTT who redecided the matter, might reach the same decision the FTT did even when the omitted factors are taken into account.

Under s12(2) of the Tribunal Courts and Enforcement Act, where a decision of the FTT is set aside we must either remit the case to the FTT with directions for its reconsideration or re-make the decision ourselves. The appellant submits both options are open to us whereas. HMRC say it should be remitted to the FTT for a full fresh decision.

We consider, taking account that given the nature of the errors means that further detailed findings of fact may need to be made, that it is appropriate for that task to be undertaken by the FTT. However, acknowledging the effort and resource that has been expended in hearing the evidence, that the FTT addressed a number of issues regarding the scope of evidence before it, which were not challenged, and that a number of primary findings made by the FTT were not subject to challenge, we consider our directions should ensure the scope of the re-hearing is commensurate with rectifying the errors and should avoid the cost and expense of a full re-hearing.

We consider the matter should (as both parties acknowledged) take place before a new panel. This does not stem in any way from any criticism of the judge but is so as to avoid any concern that a dispassionate observer would consider the panel had been subconsciously influenced by its earlier decision (see Revive Corporation Limited v HMRC [2020] UKUT 320 (TCC) (at [42]).

We will therefore remit the matter back to the FTT with the following directions for determination.

(1)

The remitted appeal must be heard by a differently constituted Tribunal (to be selected by the FTT President)

(2)

The FTT shall (subject to (4)) make its determination on the basis solely of the evidence that was taken account of by the original FTT.

(3)

The FTT shall accord Mr Galbraith’s evidence the same value as the original FTT on the basis explained at [131].

(4)

The new FTT may, as it sees fit, conduct its own test of taste and texture of the products insofar as samples of these are still available (it is understood that some are no longer in production).

(5)

The FTT may, if it remains possible, and if the parties reach agreement on bearing the cost, have recourse to the recording and/or transcript of the original FTT hearing.

(6)

The new FTT shall take as given the following primary findings of fact, but may as it sees fit make additional findings of fact on the basis of the evidence that was before the original FTT:

(a)

Organix bars: Ingredients and Process [111] – [115], The market and marketing [116] – [121], The Packaging [122] – [126], The Purchasers [131].

(b)

Nakd bars: Ingredients and Processing [186] to [191], Packaging [192] – [194], Positioning in store [195] – [196], On-line marketing [197] – [198], Purchasers [201] –[202].

(7)

The FTT shall, having made any such additional findings of fact on the basis of the evidence that was before the original FTT (and if applicable its own taste test of the product samples) perform its own fresh evaluation of whether the products are “confectionery”.

(8)

The FTT shall make whatever directions it sees fit regarding the format of the hearing such as the manner and timing of submissions from the parties on the significance of the factors the original FTT omitted, the additional findings to be made, and the issues before the tribunal including, where appropriate issues concerning quantum.

Before the FTT the parties raised the disputed issue regarding the quantum of the output tax to be repaid. The appellant contested HMRC’s position that only the net amount should be repaid (after deducting the input tax the appellant paid to the product manufacturers when it bought the products from them), arguing all of the output VAT should be repaid. The FTT did not deal with that argument as it only became relevant if the appellant were successful.

Pending a determination on remittal to the FTT, the issue is similarly academic before us. We decline HMRC’s suggestion that we should deal with the legal issues raised by the dispute at this point. The issue may not even require resolution so any determination we would make would be on an obiter basis. If the issue does require resolution, because the appellant is successful before the FTT, we consider the quantum matter would best be resolved in tandem with the FTT’s substantive decision, so that if there is any onward appeal both issues may be dealt with together.

The issue of quantum will accordingly be dealt with by the FTT upon hearing submissions either at the substantive hearing, or subsequently in the event the appellant is successful, as the FTT sees fit.

The appellant’s appeal is allowed. The appeal is remitted for a decision by a new FTT panel in accordance with our directions above at [143].