Innocent Ltd v Revenue & Customs [2010] UKFTT 516 (TC)

FTT-Tax

Innocent Ltd v Revenue & Customs

[2010] UKFTT 516 (TC) · 2010-10-25

[1]The Appellant appeals against a decision of HMRC contained in their letter dated 26 April 2007 that the pure fruit smoothies produced by the Appellant for VAT purposes fell to be standard rated for VAT purposes as beverages under excepted item 4 of Group 1 of Schedule 8 of the Value Added Tax Act 1994 (“VATA”). Products in issue

[2]HMRC’s decision was prompted by a voluntary disclosure made by Innocent on 12 April 2007 seeking to recover VAT accounted for on sales of their fruit smoothies. In the normal course of its business, Innocent changes the recipes it uses and also discontinues some smoothies and introduces new ones. It also has a guest smoothie in production for 3 months at a time. It was therefore only able to produce at the hearing samples of the smoothies in current production and our physical consideration of Innocent’s product necessarily had to be limited to these.

[3]Both parties were happy that our Decision would treat the current smoothies as representative of Innocent’s fruit smoothies which were the subject of the original voluntary disclosure. There is no suggestion that Innocent’s existing range is materially different to the range in production for the periods covered by the original voluntary disclosure.

[4]The range at the time of the hearing comprised 8 smoothies including 1 guest smoothie. The ingredients for these smoothies are set out below in paragraph 92.

[5]Innocent also produce yoghurt-based smoothies (sometimes referred to as thickies) but these were not part of the voluntary disclosure and are not part of the appeal as HMRC accept that they are zero rated as a preparation of milk (item 6 of the items overriding the exceptions in Group 1 of Schedule 8 as set out in paragraph 7 below).

[6]Innocent’s voluntary disclosure included a claim for the VAT accounted for on their range of children’s smoothies. At the hearing Innocent made it clear that they did not wish us to consider the children’s smoothies although they were part of the original voluntary disclosure. Innocent now accept that their children’s smoothies are properly standard rated. Legislation

[7]The legislation with which this case is concerned is short. It is contained in Group 1 of Schedule 8 VATA. It is the food group. It sets out what is to be zero rated as a food and says in so far as relevant to this appeal:

“The supply of anything comprised in the general items set out below, except- (a) …. (b) a supply of anything comprised in any of the excepted items set out below, unless it is also comprised in any of the items overriding the exceptions set out below which relates to that excepted item. Item No 1 Food of a kind used for human consumption. 2 Animal feeding stuffs. 3 Seeds or other means of propagation of plants comprised in item 1 or 2. 4 Live animals of a kind generally used as, or yielding or producing, food for human consumption. Excepted items 1 Ice cream, ice lollies, frozen yoghurt, water ices and similar frozen products, and prepared mixes and powders for making such products. 2 Confectionary, not including cakes or biscuits other than biscuits wholly or partly covered with chocolate or some product similar in taste and appearance. 3 Beverages chargeable with any duty of excise specifically charged on spirits, beer, wine or made-wine and preparations thereof. 4 Other beverages (including fruit juices and bottled waters) and syrups, concentrates, essences, powders, crystals or other products for the preparation of beverages. 5 Any of the following when packaged for human consumption without further preparation, namely, potato crisps, potato sticks, potato puffs, and similar products made from the potato, or from potato flour, or from potato starch, and savoury food products obtained by the swelling of cereals or cereal products; and salted or roasted nuts other than nuts in shell……. Items overriding the exceptions 1 Yoghurt unsuitable for immediate consumption when frozen. 2 Drained cherries. 3 Candied peels. 4 Tea, maté, herbal teas and similar products, and preparations and extracts thereof. 5 Cocoa, coffee and chicory and other roasted coffee substitutes, and preparations and extracts thereof. 6 Milk and preparations and extracts thereof. 7 Preparations and extracts of meat, yeast or egg. NOTES (1) “Food” includes drink. (2) ….. (3) ….. (4) Item 1 of the items overriding the exceptions relates to item 1 of the excepted items. (5) Items 2 and 3 of the items overriding the exceptions relate to item 2 of the excepted items; and for the purposes of item 2 of the excepted items “confectionary” includes chocolates, sweets and biscuits; drained, glacé or crystallised fruits; and any item of sweetened prepared food which is normally eaten with the fingers. (6) Items 4 to 6 of the items overriding the exceptions relate to item 4 of the excepted items. (7) ….”

[8]The scheme of this legislation is that food is zero rated. Food includes drink (see note (1)). Excepted items, however, are not zero rated despite being food. An excepted item is a beverage (see Excepted item 4).

[9]It was accepted by both parties that all beverages are food: indeed unless they are food they could not be within the zero rating of Group 1. So the question for this Tribunal is not whether fruit smoothies are food: all parties accept that they are and we agree. Both parties were also agreed that the smoothies are a drink in the sense that they are intended to be drunk. The question for the Tribunal is whether the smoothies are a beverage. The Appellant’s case is that fruit smoothies are merely a liquid food and not a beverage: HMRC considers them to be a beverage.

[10]Various items override the exceptions but it was not suggested that a fruit smoothie fell into any of the relevant overrides. The relevant overrides are items 4 to 6 which include tea, maté, herbal tea, cocoa, coffee chicory and similar items, and milk and preparations of milk. What is a beverage? Social policy

[11]The lawfulness of the UK’s zero rating provisions were not in issue (as we would expect bearing in mind that one party was HMRC and the other was seeking to rely on zero rating). Mr Cordara suggested however that the EU provision which permits the UK to zero rate requires our legislation to have a social policy and that we should therefore find a social policy in Group 1 and interpret the meaning of “beverage” accordingly.

[12]There are really two questions here. Is “beverage” to be interpreted according to social policy and if so, what is that social policy?

[13]The Principal VAT Directive 2006/112/EC at Article 110 provides, as did Article 28(2)(a) of the Sixth VAT Directive before it, that:

“Member States which, at 1 January 1991, were granting exemptions with deductibility of the VAT paid at the preceding stage or applying reduced rates lower than the minimum laid down in Article 99 may continue to grant those exemptions or apply those reduced rates. The exemptions and reduced rates referred to in the first paragraph must be in accordance with Community law and must have been adopted for clearly defined social reasons and for the benefit of the final consumer.”

[14]Mr Codara thinks that there must be a social policy behind the UK’s zero rating of food and that this social policy must be for clearly defined social reasons and for the benefit of the final consumer. He cited the Court of Appeal decision in Jubilee Hall Recreation Centre

[1999] STC 381

at page 385 where the Sir John Vinelott said of Schedule 8 (the zero rating schedule) in relation to Article 110 that it “…must be construed in a way which gives effect to this manifest purpose, even though this may involve some departure from the strict and literal application of the words used or the implication of words necessary to give effect to that purpose.”

[15]To be lawful, the food zero rating provisions in UK domestic legislation must be for clearly defined social reasons and for the benefit of the final consumer: and removing food from the charge to tax would clearly seem to fulfil these requirements. But what of the exceptions to the food zero rating, such as the exception for beverages? Mr Cordara says that these exceptions must also be part of the social policy: HMRC does not agree.

[16]We find that this question has been considered before. In the High Court decision in Proctor & Gamble

[2008] STC 2650

at paragraphs 45-48 Warren J considered this issue in relation to Regular Pringles, the product at issue in that case, and said that although UK VAT legislation has to be construed with the social policy in mind, this can apply only where the policy can be established. Further:

“The starting point is EU legislation which allows, on grounds of social policy, for the zero-rating of supplies. It is on this basis that the UK legislation has an exemption for food. There is, however, no requirement that any exclusions from exemption for certain types of food be justified by any social policy. It does not seem to me, therefore, that any policy in relation to the exception of items from zero-rating is to be found in the EU legislation.”

[17]Although this decision of the High Court was appealed, it does not appear that this point was appealed. The Court of Appeal commented at paragraph 5 of the decision that it was by then “common ground” between the parties that Article 110 permitted but did not require zero rating and provided no aid to construction in the case. We are therefore bound by the views of the High Court that the exceptions to zero rating do not need to be governed by social policy.

[18]Although of no legal relevance, we note that in the case of EC Commission v UK C-416/85

[1988] STC 456

, where the EC Commission challenged a great many of the UK’s zero rates as lacking in clearly defined social purpose or not being for the benefit of the final consumer, they did not challenge the zero rating of food for human consumption. They did (unsuccessfully) challenge the other items of group 1 (ie items 2-4, the zero rating of live animals for human consumption, feeding stuffs for such animals and seeds). So, implicitly, the opinion of the EC Commission was that the UK’s zero rating of food for human consumption was lawful.

[19]If, on the contrary, we had concluded that we did have to find a social policy that not only justified the zero rating of food but the exceptions from it, and interpret “beverage” accordingly, we would be in difficulties. We are not able to find a clear policy behind the exceptions.

[20]In the EC Commission v UK case the UK Government stated that the social policy behind zero rating items 2-4 in group 1 was to keep food prices down (see pages 462 and 473). It seems an obvious inference in any event that the entire purpose of Group 1, especially item 1, is to keep down the cost of food.

[21]But what is the social policy behind the exceptions? There have been suggestions in previous cases, such as put by counsel in the High Court hearing in Kalron Foods Ltd

[2007] STC 1100

at page 1103, that there is an anti-junk food policy in Group 1. As Warren J said, the legislation does not bear this out. We are bound by Warren J’s views and in any event would come to the same conclusion. Cakes and many biscuits are zero rated while fruit juice and bottled water are standard rated. We consider it general knowledge that the latter are considered better for you than the former. Further, many processed foods are zero rated and yet would no doubt be considered junk food by many nutritionists.

[22]Mr Cordara suggests that the social policy is that the UK zero rates foods with nutritional value. He points out that Innocent’s smoothies are treated by the Government as equivalent to 2 out of the 5 recommended daily portions of fruit and vegetables. He considers that the main purpose of a beverage is not to provide nutrition but – as per the test in Bioconcepts Ltd (1993) VTD 11287 (see paragraph 41 below) – to increase bodily liquid levels, to slake the thirst, to fortify, or to give pleasure. So, he says, it follows that sugary “soft” drinks are beverages but smoothies are not. In his view, a smoothie is like a liquefied fruit salad and, as expert evidence shows, full of nutrition.

[23]HMRC do not suggest that there is a social policy for the exclusions: in their view there is only a social policy for the headline zero rating for food.

[24]We cannot agree with Mr Cordara that there is a clear social policy for the standard rating of beverages. It is not to standard rate “junk drinks”. Although alcoholic drinks and soft drinks which we think many people would regard as not good for you are standard rated, fruit juice and bottled water are also expressly standard rated and we would not regard them as junk drinks. We also do not agree that there is a social policy to zero rate only nutritional drinks. Sugar is a nutrient. Plenty of beverages containing sugars are standard rated, such as soft drinks and fruit juices. It is not even to zero rate only drinks which contain nutrients other than sugar. If there was such a policy tea would not be zero rated. It may be well be good for you but (unless milk or sugar has been added) would not normally be regarded as nutritional. We are in any event bound by the views of Warren J in Kalron that there is no social policy for the beverage exception, see in particular paragraphs 9, 10 & 11 at page 1103-4.

[25]Mr Cordara then suggested that the social policy is that drinkable foods are intended to be zero rated, and suggests that if the principal function of a liquid is nutrition and not stimulation, refreshment, or amusement (or presumably hydration) then it should be zero rated as a liquid food. We agree that the legislation does make a distinction between drinkable liquids and beverages: if this were not the case it would be “liquids” rather than “beverages” within the excepted items, and soup would be standard rated rather than, as all parties agree, zero rated. But this is really the same as asking us to decide what is a beverage (rather than a drinkable liquid) and we will return to this point below. We note however that this distinction is not a consistent social policy, because some beverages (such as tea and coffee) are zero rated despite not being liquid foods.

[26]In conclusion we have been unable to identify any consistent social policy behind the exception of certain beverages from the food zero rating. Neither side is suggesting that the UK’s food zero rating is actually unlawful: to win such an argument would not profit the Appellant. The High Court has ruled in Proctor & Gamble as set out above that HMRC do not need to justify exceptions from zero rating with social policy and in any event the Appellant has not been able to suggest a coherent social policy for the beverage exception in the legislation. We find no help in Article 110 in resolving the question before us which is whether Innocent’s fruit smoothies are beverages. Ordinary meaning of beverage?

[27]As we find no help in a social policy in interpreting the excepted item 4 “other beverages” and whether fruit smoothies are within it, we need to decide what is the meaning of “beverage”. It seems to us that beverage must have its ordinary meaning. This is because its use in the legislation does not suggest that it carries anything other than its ordinary meaning, and there is authority from the higher courts (see paragraph 29 below) that the zero rating provisions are indeed to carry their ordinary meaning.

[28]Looking at the use of the word “beverages” in Group 1 we find it refers to alcoholic beverages in excepted item 3 and “other beverages” in excepted item 4. Those “other beverages” are stated to include fruit juices and bottled water, and (from the overriding items) tea, maté, herbal tea, cocoa, coffee and chicory, and milk. All these drinks are what this Tribunal would understand ordinarily to be within the meaning of the word “beverages”. The one possible exception to this is milk, which although frequently drunk as a drink is also frequently consumed by being eaten and not drunk when it has been used in cooking or served or eaten with a spoon on breakfast cereal. Most likely it was put into the overriding items because there was doubt whether it was a beverage and the drafters wished to ensure its zero rated status was not in doubt.

[29]The Ferrero UK Ltd

[1997] EWCA Civ 1623

[1997] STC 881

case was about whether certain wafers were zero rated as biscuits, as excepted item 2 (see above) excludes from zero rating confectionary except cakes and non-chocolate covered biscuits. The Court of Appeal agreed with the Tribunal that the question was whether the wafers were biscuits. It agreed with the Tribunal’s statement that “The words in the statute must be given their ordinary meaning. What is relevant is the view of the ordinary reasonable man in the street”. The Court of Appeal [paragraph 884g] explained this as meaning:

“That is, what is the view of the ordinary person as to the nature of the product and whether or not the product is one which falls within the relevant category which here is that of a biscuit.”

[30]The Court of Appeal again considered the interpretation of Group 1 in Proctor & Gamble (UK)

[2009] EWCA Civ 407

[2009] STC 1990

, which concerned Regular Pringles, which this Tribunal would describe as a kind of processed crisp. The question was whether it was ineligible for zero rating because it fell into the excepted category 5 in that it was a product similar to a potato crisp and “made from the potato”. Its main ingredient, of which the final product comprised about 42%, was potato. The Court of Appeal gave guidance on how the question should be approached. Jacob LJ at paragraph 35 said:

“The statute is simply posing a kind of jury question ‘is it similar to a potato crisp etc and made of potato? The question is not capable of elaboration or complex analysis.”

[31]Toulson LJ said (referring to the Court of Appeal’s decision in Ferrero ) at paragraph 63:

“I rather regret the introduction of the ordinary man in the street into this area because I do not regard it as necessary and it has led on to a distracting argument about what knowledge should be attributed to that hypothetical person. I agree with Jacob LJ that the approach approved in Ferrero 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. In determining that question I do not see any advantage is gained by referring to the hypothetical ordinary person in the street.”

[32]Mummery LJ at paragraph 79 said that:

“…it is vital to recall why the tribunal was required in the first place to answer the question whether the goods in question are ‘made from’ the potato. It was not in answer to a scientific or technical question about the composition of Regular Pringles, or in response to a request for a recipe. It was for the purpose of deciding whether the goods are entitled to zero-rating. On this point the VAT legislation uses everyday English words, which ought to be interpreted in a sensible way according to their ordinary and natural meaning. The ‘made from’ question would probably be answered in a more relevant and sensible way by a child consumer of crisps than by a food scientist or culinary pedant…”

[33]Warren J in the case of Kalron, where the meaning of “beverage” was under consideration also concluded that (paragraph 42):

“…it was for the Tribunal to determine whether the Product is a beverage or not applying the ordinary meaning of that word as a matter of the English language unless the context of the VAT legislation requires a special meaning to be attributed to it. In saying that, I am aware that words often have slightly different meanings to different people. There may be no unique “ordinary” meaning of a word, particularly of a word such as beverage which is not normally used at all in modern times in everyday speech….”

[34]He went on to conclude (in paragraph 68 which is set out below) that the word beverage in Group 1 did not have a special meaning. Warren J’s decision on a matter of law is binding on us.

[35]Our conclusion is that ‘beverage’ must be given its ordinary and natural meaning, and the Tribunal must decide whether a fruit smoothie is within that ordinary and natural meaning of the word beverage based on the evidence in front of it. Ordinary and natural meaning of beverage

[36]What is the ordinary and natural meaning of beverage? The Oxford English Dictionary puts the main meaning as: Drink, liquor for drinking; esp. a liquor which constitutes a common article of consumption.

[37]As Warren J noted in Kalron , the word beverage is not commonly used now. There is also the point that smoothies are a recent innovation. Would the draftsman of Group 1 back in 1972 have considered a smoothie to be a beverage? If asked “is a smoothie a beverage?” we assume he would answer “what’s a smoothie?”. Yet if we were now in the year 2010 to follow Mummery LJ’s suggestion (see paragraph 32 above) of asking a child the same question, his reply would no doubt be “what’s a beverage?”.

[38]The dictionary suggests that all drinks are beverages and HMRC’s view is that, therefore, the meaning of beverage is the same or virtually the same as “drink”. The Appellant does not agree. In any event, as has been noted in the cases, that does not mean that all things that can be drunk are drinks. In Kalron, Warren J agrees with the statement of Sir Stephen Oliver in Bioconcepts that not all drinkable liquids are beverages but goes on to say that “some caution must be exercised in placing any reliance on any supposed distinction between drinks and beverages” (paragraph 58). Warren J appears to see the real distinction as being between drinks and drinkable liquids. He sees the dietary integrator such as in SIS (Science in Sport) Ltd (2000) VTD 16555 which was made up in liquid form for consumption as being a drinkable liquid but not a drink: [paragraph 68 on page 1115] “Mr Thomas [counsel in Kalron ] says that [the word beverage] must have a special meaning because the dictionary definition equates it with ‘drink’ but there are drinks which are clearly not beverages for VAT purposes. That argument is, I consider, based on confusing two different meanings of the word drink. The contrast which I would draw, and which I have mentioned already and which is referred to in SiS [2000] V&DR 195, is between a drink within the ordinary meaning of that word and a drinkable liquid, that is to say a liquid which is fit for human consumption and which, being liquid, is consumed by drinking it. The dictionary definition is looking at the ordinary meaning of the word drink whereas the cases which appear to identify drinks which are not beverages are dealing with liquids which would probably not be called drinks, within the ordinary meaning of that word, either. But even if that is wrong, so that there are drinks ordinarily so called which are not beverages, that will be so, in my view, only if the ordinary meaning of the word beverage excludes that drink. As to that, it may be that milk, for example, would not, ordinary speaking, be regarded as a beverage even if it is a drink. Accordingly, I reject Mr Thomas’ submission that beverage has a special meaning in Group 1 different from its meaning as a matter of ordinary language.”

[39]The Appellant accepts that fruit smoothies are drinks and points out that Warren J did not say that ‘drink’ was synonymous with beverage. It relies on the definition of beverage in Bioconcepts and considers that a fruit smoothie does not meet it. Mr Cordara says that to refer to “a drink of milk” is correct use of language but milk is not a beverage so not all drinks are beverages even if they are not merely drinkable liquids. As mentioned above, see paragraph 28 and the quotation from Warren J at paragraph 38, there is some dispute whether milk is a beverage: it is specifically exempted from standard rating as a beverage so clearly the draftsman thought it could be seen as a beverage and that he needed to ensure its zero rated status was beyond dispute. So it is not obvious to the Tribunal that milk is not a beverage and therefore we cannot draw the conclusion that Mr Cordara asks us to that because milk is not a beverage (he says) there must be drinks which are not beverages. It may be that milk is either a beverage or in a special category of being both a drinkable food that is not a drink (like soup) and a beverage. But we are not asked to decide this and draw no conclusions.

[40]We consider following Warren J, that the meaning of “drink” is very close to that of beverage and does not encompass merely drinkable liquids. We need to decide whether a fruit smoothie is only a drinkable liquid, such as food in drinkable form, or whether it is a beverage. Is it drunk because it is a drink? If so, it is likely to be a beverage. The Bioconcepts test

[41]Bioconcepts Limited VTD 11287 was a decision of the VAT & Duties Tribunal in 1993. It has been cited with approval so far as we can see in every tribunal case on beverages since then. It considered whether a food supplement was a beverage. The supplement was intended as a slimming aid which was to be sipped throughout the day and would be a violent laxative if a glassful of it were drunk at once. Sir Stephen Oliver considered the meaning of “beverage”:

“It seems to us that notwithstanding the Oxford English Dictionary of “beverage” meaning drink, it is not used in the sense of meaning all drinkable liquids. Its meaning in ordinary usage covers drinks or “liquors” that are commonly consumed. This is the primary meaning in the Oxford English Dictionary. Liquids that are commonly consumed are those that are characteristically taken to increase bodily liquid levels, to slake the thirst, to fortify, or to give pleasure. That meaning covers the liquids recognised [by counsel] as beverages (eg alcoholic liquids, tea, coffee, cocoa, chocolate, and soft drinks and meat-based preparations.)”

[42]Sir Stephen considered, as Warren J has since ruled, that ‘beverage’ did not include all drinkable liquids. He considered that it did include those drinks that were commonly drunk to hydrate or thirst-quench, to fortify or give pleasure.

[43]Warren J considered the Bioconcepts test in Kalron and did not suggest it was wrong but did say at paragraph 60:

“I would be surprised if the tribunal had thought that it was laying down an exhaustive definition of what a beverage is rather than listing common characteristics of a beverage not all of which needed to be present in any particular case. I do not consider that the tribunal can be taken to have ruled out other drinks not having any of those characteristics from being considered a beverage.”

[44]We therefore agree with Ms Mitrophanous that the Bioconcepts test is helpful but it is neither cumulative nor exhaustive. And indeed we do not think that Sir Stephen thought that it was.

[45]What does the test mean? We note that the Bioconcepts test refers to whether the drink is “commonly consumed”. We understand this merely to mean whether a common reason for its consumption is one of the four reasons given by Sir Stephen. We do not understand it to mean that to be a beverage the drink must be consumed by a large section of the population of the country (and neither party did suggest that this is what it meant) nor do we find that this is part of the ordinary meaning of beverage.

[46]The meaning of the first two parts of the Bioconcepts test – whether the drink is commonly consumed to increase bodily liquid levels or to slake the thirst – did not seem to pose any problems. The dispute between the parties on this was whether fruit smoothies as a matter of fact were commonly consumed to hydrate or quench thirst.

[47]There was dispute as to the meaning of “fortify”. Mr Cordara equated “fortify” with instant gratification: the sort of instant gratification, as he said, as one could find in a hip flask. We agree that fortification in this sense covers alcohol, which is commonly administered to revive after a shock or to give Dutch courage. Counsel for HMRC did not consider the meaning of “fortify” to be so limited. She saw smoothies as fortifying because they are good for you. We agree that stimulants such as tea and coffee (clearly beverages) could be seen as fortifying. We can also see that sugary soft drinks might be seen as “fortifying” in the sense of giving a sugar rush and immediate energy (which seemed to be a meaning given to it in Alpro (UK) Ltd (2006) VTD 19911). However, we don’t think Sir Stephen meant “fortifying” in the context of the normal “fortification” to be obtained from eating all foods, and we think it would be nonsensical to interpret it in this way as it would fail to distinguish between beverages and food.

[48]The meaning of the last test from Bioconcepts is also not particularly clear. When are drinks commonly consumed to give pleasure? Clearly the consumption of many if not most foods and drinks will commonly give pleasure. It seems to us Sir Stephen had in mind the consumption of a drink that was drunk principally for pleasure. This might, for instance, encompass alcoholic and soft drinks. The kind of drink you would offer a guest

[49]We asked Ms Mitrophanous for HMRC if HMRC had a definition of beverage. They rely on the Bioconcepts test in their public notice but consider that it is not exhaustive. Ms Mitrophanous did not suggest that HMRC had an exhaustive test but said she considered that a “good working description” would be that a beverage is the kind of drink that one might commonly offer a guest. A soup would not be offered as a drink. Liquid medicine or a dietary supplement would not be offered to a guest as a drink. But she thinks, if the host had some in the house, he would offer a smoothie to a guest as a drink.

[50]The Appellant’s view was that this was not the correct test and indeed so far as we can see it is not as such the test employed before in the Tribunals or the courts.

[51]We think that it is not the definitive test for a beverage (and HMRC did not suggest it was) but that it is something that we would take into account. It is more likely that a smoothie is a beverage if it is the sort of thing that would be offered to an unexpected guest as a drink. Indeed, Warren J at paragraph 78 of Kalron indicates that the Tribunal must consider the circumstances in which the drink in question is drunk: is it drunk in the same circumstances as one might consume a beverage? So it is relevant but not conclusive whether it would be offered to a guest as a drink. Technical definition from Professor Strain

[52]We had expert evidence from Professor Sean Strain, and we consider his evidence in more detail below when we consider the facts in this case. On the question of the meaning of “beverage” he said that although there is no scientific definition of beverage, he considers that a food is anything that makes the digestive tract work and a beverage (or drink) is anything that does not.

[53]He says that the purpose of the digestive tract is to absorb nutrients from the environment. Some nutrients, such as sugars, can be absorbed without digestion. Others cannot be absorbed unless digestion takes place. The digestive tract is therefore a complicated organ as it has to digest proteins and complex carbohydrates in order to allow these nutrients to be absorbed into the body. Sugars, on the other hand are either absorbed directly (in the case of glucose) or with the help of an enzyme (in the case of sucrose). Sugars are not digested.

[54]Fruit contains pectin and other similar soluble fibres which work by slowing the upper digestive tract and make the consumer feel full. When they reach the lower tract the fibre is digested by bacteria (using enzymes humans do not have) which release nutrients as by-products, which humans can absorb.

[55]Fruit juice at best contains trace fibre which is insufficient to have an effect on the digestive system. The sugars contained in fruit juice and soft drinks are absorbed directly without digestion.

[56]By Professor Strain’s test milk is not a beverage as it contains protein which causes the digestive tract to secrete enzymes in order to digest it.

[57]Mr Cordara in his closing said that the Appellant supported Professor Strain’s definition and that they considered it explained the existing case law. In support of the Tribunal using a scientific test for “beverage”, Mr Cordara pointed to the HMRC Public Notice 701/14 Food of May 2002 which contains what he considers to be quite a scientific test for whether a sports drink is a beverage. At paragraph 4.6.1 HMRC state that a sports drink is standard rated unless it meets a list of criteria including that it has as its main ingredient carbohydrate (other than sugar), creatine, protein, or a mixture of these. Mr Cordara says that to answer this would require scientific analysis of the drinks.

[58]We do not accept Professor Strain’s test as definitive. Firstly, it does not describe what this Tribunal considers to be ordinarily understood by the meaning of the word “beverage”. We do not think when considering what is a beverage a person would stop to consider how the digestive system operated on the liquid consumed.

[59]Secondly, we do not think it is what Parliament had in mind when Group 1 was drafted else they would have incorporated this explanation into the legislation.

[60]Thirdly, the Court of Appeal, in Proctor & Gamble (above) a ruling which is binding on us, has said that the test should not be scientific. Although they were clear (witness Toulson LJ’s comment at paragraph 31 above) that the Tribunal is not limited to the knowledge of the ordinary man in the street. From this it follows that we can take note of the expert evidence that was given to us about the nutritional composition of a smoothie and its effect on the digestive system. Nevertheless they were also clear that the test we use the evidence to answer is not a scientific one (see for instance Toulson LJ’s comment cited at paragraph 31 above: “…it conveys that this is not a scientific question.”) We note that although HMRC might choose to use at least in part a scientific test to describe a sports drink that is not a beverage, this test pre-dated the decision in Proctor & Gamble and is in any event only HMRC’s view of the law and not binding on us.

[61]Fourthly, we do not accept that the test describes the differences between what we know from the legislation are beverages. Professor Strain’s evidence was that milk does require digestion as it is protein yet it seems milk is perhaps properly considered as having a double identify as a beverage and a food that is eaten (see paragraphs 168-170). Even putting aside the status of milk, hot tea containing milk or cocoa made up with milk would properly we think be described as a beverage yet, from Professor Strain’s evidence will require some digestion. The same it seems is true of untreated cloudy apple juice which we were told by Mr Reed would contain similar amounts of soluble dietary fibre to smoothies, and therefore will, according to Professor Strain, require digestion.

[62]Fifthly, the test does not explain the case law: while it is consistent with the finding in Alpro that soya milk was not a beverage it is not consistent with the finding in Kalron that a drink (also a smoothie) made of all the edible parts of fruit and vegetable was a beverage as on Professor Strain’s evidence that drink would require digestion.

[63]Nevertheless, we do consider that a factor that we should take into account in deciding whether a smoothie is a beverage would be that it does, as Professor Strain described, have an impact on the digestive system whereas many (but not all) beverages do not. The Kalron decision

[64]It was HMRC’s view, as expressed in its decision letter dated 26 April 2007, that the Kalron decision is a binding ruling by the High Court that all smoothies (other than those based on milk) were standard rated beverages. Ms Mitrophanous at the hearing pointed out that it would be anomalous if we reached the conclusion that Innocent smoothies were zero rated when the High Court had ruled Zumo smoothies were standard rated.

[65]The facts found by the Tribunal in Kalron Foods Ltd (2006) VTD 19738 were that the taxpayer made smoothies fresh on the customer’s order from the liquefied edible parts of fresh fruit and vegetables. The resultant drink was served in disposable take away cups at Zumo bars. Some bars had tables and chairs for customers but others did not. The Tribunal found that “although quite thick, the products were not difficult to drink. The colour was not such to be off-putting. The result was a refreshingly healthy drink.”

[66]The Tribunal had evidence on how the smoothies were produced and the context of the bars in which they were sold but no other evidence was put to the Tribunal. The Tribunal decided the case against the taxpayer. In particular, the evidence led on how the smoothies were sold at special Zumo bars did not show to the Tribunal whether the smoothies were being purchased as a substitute for a different beverage (say something alcoholic) or being purchased as a food substitute (see paragraphs 34 & 35). Either would be possible and the Appellant had failed to bring evidence which showed that it was not a beverage. It is clear from paragraph 33 that the Tribunal did not think that the nutritional value of the smoothie was decisive but that the question was whether they were consumed as beverages.

[67]The case was appealed to the High Court. Warren J decided that on the basis of the evidence in front of it the Tribunal had come to a reasonable conclusion on the facts. Furthermore it was clear that Warren J considered that the Tribunal had come to the right conclusion and the one he himself would have reached: “There are sufficient findings of fact for me, on appeal, to decide whether or not the Products are beverages. I would conclude, taking into account all of the factors which Mr Thomas [counsel for the Appellant] submits are relevant, that they are.” (paragraph 82).

[68]Mr Cordara does not consider the Kalron case to make a binding finding on this Tribunal that fruit smoothies are standard rated beverages. HMRC thinks the Kalron case is binding on us unless the products are differentiated. We agree with Mr Cordara as far as it goes: Tribunals make their decisions on the facts in front of them. In this case, the evidence is not the same. The products are not identical in composition nor in how they are marketed and consumed (as explained below). Further, we have had a great deal of evidence on marketing and on the nutritional effect of smoothies which was not led in the Kalron case. We see it as open to us on the basis of this evidence (if the evidence warrants it) to reach a different conclusion on the facts. Nevertheless, Warren J’s approach to the law is binding on us and it is therefore a highly relevant authority for us considering as it does the meaning of “beverage” in the context of smoothies.

[69]Warren J at paragraph 75 said that the process of blending or liquefying food could be enough to convert it from food to a beverage. He also ruled that the Tribunal is entitled to and ought to take into account the circumstances in which the product is consumed. We take this as meaning, as we have said above, that we should consider whether the smoothies are consumed as a beverage. He agreed with counsel for the Appellant that the Tribunal ought to consider questions such as the ingredients, the nutritional value, the manufacturing process, the place of sale, the reason for consumption, and the appearance and texture of the product. But he also said at paragraph 73 that these questions were neither exhaustive nor definitive. When considering the reasons for consumption, whether as an alcohol or as a food substitute, he put emphasis on how the product is drunk:

“What might be more significant, however, in each example is not what the Products substitutes for , but how it is taken: in each case it is drunk and consumed as a drink.”

[70]In saying this Warren J cannot have been referring merely to the fact that the product is drunk. He has already made the distinction (to which we refer in paragraph 38 above) between a drink (including a beverage) and a drinkable liquid. So “how it is taken” and being “consumed as a drink” must mean more than merely it is drunk. Putting aside for the moment that he recognised that “beverage” might have a slightly narrower meaning than “drink” in this context, his distinction was between drinks and drinkable liquids. What distinguishes them is that the former is drunk because it is a drink but the drinkable liquid is consumed for other reasons. The Bioconcepts test recognises that liquids are drunk as beverages if the reason is to hydrate, quench thirst, fortify or give pleasure. Warren J recognised that they may be other reasons to drink beverages, but that to be a beverage the drink must be consumed as a drink – which we take to mean also because they are a drink – in order to be a beverage. Conclusions on the law

[71]We conclude that when deciding whether Innocent’s fruit smoothies are beverages we should consider whether they are within the ordinary meaning of beverage.

[72]We also bear in mind that in answering the question of whether a smoothie is a beverage we need to consider what other drinks are or are not beverages. Warren J said in Kalron at page 1106 “…it is not right simply to ignore other products which clearly are beverages (such as fruit juices) to see what characteristics there may be in common and which might point to the Products being beverages”.

[73]But we also consider: · How the smoothies are consumed. Are they consumed as drinks or are they merely drinkable liquids? · how the smoothies are made and from what, and what is their resulting appearance and texture; · where the smoothies are sold and consumed; · when the smoothies are consumed; · their effect on the human body; and · why the smoothie is consumed. Evidence Witnesses

[74]We heard evidence from Mr Richard Reed, a co-founder of Innocent, and currently brand director and co-CEO of the Company. His role is to take responsibility for the strategic direction of the Company, to decide which new products to launch and to be responsible for the Innocent brand image.

[75]We found him to be a good witness with a thorough knowledge of the various marketing research which was produced by Innocent in evidence and on which he was cross-examined in great detail. Counsel for HMRC did point out that, judging by a chart put in evidence by Innocent after Mr Reed wrote his witness statement, he was mistaken in saying that August was the month of lowest sales. The chart showed that (as he said) January was the month of highest sales, but the chart also showed that sales in August were middling with December actually being the month of lowest sales. However, we found that the witness statement had not been made with the specific chart in mind, and what Mr Reed said may have been true for earlier years. In any event the mistake, if it was one, was not of the sort that affected our overall impression of him as a good witness. The expert witnesses

[76]We had expert evidence from Professor Strain appointed by the Appellant and Miss Gearey appointed by HMRC. Evidence of Professor Strain

[77]Professor Sean Strain is currently co-Director of the Centre for Molecular Biosciences and a Professor of Human Nutrition at the University of Ulster. He has a BSc in Chemistry, a BAgr in Agricultural Chemistry and a PhD in Nutritional Biochemistry from Queen’s University of Belfast. His career has been as lecturer, then senior lecturer then professor in human nutrition at various Universities. He is the author of over 200 peer reviewed research publications. He is or has been a member or chair of numerous Boards concerned with human nutrition. The Tribunal has no hesitation in accepting him as an expert witness on the subject of human nutrition.

[78]We also found Professor Strain to be a good witness and we accept the evidence he gave (and which we describe below in our findings of facts). We note, however, that Counsel for HMRC was critical of his evidence so we set out here why we do not consider the criticisms justified.

[79]One criticism made by Ms Mitrophanous was that Professor Strain had contradicted himself by saying that smoothies (made from the whole fruit excluding only the inedible parts) were nutritionally identical to the raw fruit but that nevertheless elsewhere he said that eating raw fruit was healthier than eating smoothies. The Tribunal understood his explanation of this to be that nutritionally they are identical because they comprise the same elements (such as the identical amount of fibre, trace minerals etc) but nevertheless the process of liquefaction has homogenised the fruit and that for reasons not fully understood by science this disruption of the matrix of the fruit means that a smoothie is not as healthy for an individual as eating the raw fruit (although considerably more healthy than eating no fruit at all). We entirely accept his evidence on this, and do not consider that it was contradictory.

[80]Ms Mitrophanous also said that in her opinion Professor Strain was drawing conclusions from the reports he cited that were not justified particularly in relation to the effect of 3g of pectin on satiation. We explain below in paragraphs 99-121 what those conclusions were. We consider that if HMRC did not accept the conclusions Professor Strain drew, which were set out in his witness statement, it would have been more appropriate for them to have called their own expert witness in nutrition to challenge the evidence. We accept his evidence as he was an expert, there was no contradictory evidence from another expert, and we had found him to be a good witness. His evidence was also consistent with our sampling of the product.

[81]We also note that in closing Ms Mitrophanous suggested that Professor Strain had at the end of his evidence withdrawn his conclusion about the effect of 3g of pectin. This was not our understanding at the time nor on re-reading the transcript. We consider that he merely clarified, as his report says, that he considers 3g or more of pectin and/or similar soluble dietary fibre would have the effect he describes on satiation in most people.

[82]Ms Mitrophanous also pointed out that when Professor Strain gave evidence in the Tribunal case of Ocean Grown UK Ltd VTD 20562, which concerned a protein-rich drink, he had referred to smoothies as beverages. The Tribunal did not consider this detracted from his evidence: Professor Strain’s evidence was directed to the properties of smoothies and other drinks. The legal definition of a beverage is for the Tribunal. We note that his evidence in Ocean Grown (paragraph 20) is consistent with his evidence in this case, that smoothies contain a greater amount of soluble dietary fibre than fruit juices, and that they act on the small intestine. Evidence of Miss Gearey

[83]HMRC’s expert witness was Miss Gearey. She has a BSc in Applied Biological Sciences and has spent her career of some 16 years at Campden BRI. She is the Senior Technical Officer in the Department of Chemistry & Biochemistry where she has worked in the Food Specification & Control group for 15 years. For the last 13 years she has been the manager responsible for this group. Campden BRI undertakes research and testing, including on behalf of the Government. It is part of Miss Gearey’s role to analyse samples submitted to her department. Her expertise is in the characteristics, composition and chemical analysis of stuffs (often food stuffs) and devising tests to differentiate products. Much of her work is in the subjective area of sensory perception. She is not an expert on nutrition and did not give evidence about nutrition.

[84]For the purpose of this hearing she was asked to comment on the thirst quenching properties of smoothies. We found her to be a good witness and we accepted the evidence she gave and refer to it in paragraphs 151 & 155 below. Although the expert witnesses themselves did not appear to agree with each other, the differences between them seemed to us to be a matter of degree in that both witnesses accepted that smoothies had some thirst quenching properties but differed as to the extent of these properties. Market Research

[85]We were presented with the results of some market research. There was: Kantar : Innocent compiled some charts based on data produced in November 2009 by Kantar, which is an independent body which carries out market research on a national panel of some 11,000 consumers. Innocent has no influence over the questions posed by Kantar but merely buys in the research results. AC Nielsen: Innocent buys in data compiled by AC Nielsen from the bar codes of items sold in all supermarkets in the UK in a certain period. The value of the data is that it informs the manufacturer of how much of any particular product is selling in the whole of the UK or part of the UK at any particular time. Incite: Innocent compiled some charts based on market research commissioned by Innocent from Incite in June 2008. Incite asked a panel of 1,200 consumers. Innocent’s purpose was to understand the motivations of consumers consuming Innocent smoothies and it did not commission the research for this Tribunal hearing. Project Mint: this was qualitative research commissioned by Innocent in March 2010 which involved in depth interviews with 10 people and 6 accompanied shopping trips. Deep Dive: qualitative research commissioned by Innocent in April 2007 which involved a very small group of people interviewed and keeping video diaries. Tangible Shopper Research : qualitative research commissioned by Innocent in September 2008 which involved market researchers accompanying a limited number of people on shopping trips.

[86]We have referred extensively to the evidence from the market research, as indeed we were referred to it extensively at the hearing. Nevertheless, we treated it with some caution. It was apparent from the different results produced from the different pieces of research, and indeed as we would expect as a matter of common sense, that the answers received depended very much on how a question was phrased and what alternative answers were offered to the respondents to the questions. We did not find it conclusive evidence and preferred to rely on it where it was consistent with evidence from other sources.

[87]Ms Mitrophanous made the point that the Kantar research looked at all smoothies, including yoghurt-based “thickies” and smoothies produced by other manufacturers, and for that reason did not consider the Tribunal should rely on it. Although we bear the qualification in mind, as we see the market research in any event as evidence to be treated with some reservation (as explained above), we do not see the Kantar research as being without any evidential value, especially as Innocent have such a high share of the market (at least 72%) and thickies are only a small percentage of smoothies sold overall. Samples

[88]The tribunal was given a number of products to sample by both sides. We had samples of the 8 smoothies in current production. We were also given a number of comparators.

[89]The comparators produced by Innocent were: · Strawberry milk · Orange juice with bits · Orange juice without bits

[90]The comparators produced by HMRC were: · Extra juicy bits orange juice · Apple juice · Carton of pomegranate juice · Innocent’s This Water (flavoured water) · Innocent’s children’s smoothies Findings of fact Contents and preparation of smoothies Ingredients

[91]Innocent’s smoothies in general contain nothing but fruit. The exception is where certain natural but non-fruit ingredients are added to a few smoothies. The only ones of which we had examples were the pineapple, banana and coconut smoothie and the lemon, honey and ginger guest smoothie. The former contains 7% coconut milk and the latter 1% honey and 0.2% ginger. The other ingredients are all fruit.

[92]The contents of each 250ml serving of Innocent’s fruit smoothies in production at the time of the hearing were: Blackberries, strawberries and boysenberries ¾ pressed apple ½ a mashed banana 5 pressed white grapes 1½ crushed strawberries (9%) 6 crushed blackberries (9%) 4 crushed boysenberries (8%) dash of squeezed orange 9 crushed blackcurrants a squeeze of lime Fibre content: 3.3g Guavas, mangoes and goji berries 1 pressed apple 1 crushed pink guava (20%) ½ a squeezed orange ¼ of a mashed banana a chunk of crushed mango 5 squashed goji berries (5%) a squeeze of lemon Fibre content: 4.8g Kiwis, apples and limes: · ¾ pressed apple (46%) · a slice of pressed pineapple · ⅓ crushed kiwi (16%) · 5 pressed grapes · ƒ of a squeezed lime (2.5%) · A dash of spinach & nettle extract Fibre content: 3.5g Mangoes and passion fruit 1 pressed apple ½ a crushed mango ⅓ a mashed banana ⅓ a squeezed orange ⅓ a pressed passion fruit and some squeezed lime Fibre content: 3.8g Pineapples, bananas and coconut: a slice pressed gold pineapple 1 mashed banana ¼ pressed apple coconut milk a squeeze of orange a dash of lemon Fibre content: 2.5g Pomegranates, blueberries and acai: 22 pressed red and white grapes ½ a mashed banana ⅓ a pressed pomegranate ⅓ a squeezed orange 32 crushed blueberries 19 peeled acai berries and a squeeze of lemon Fibre content: 3g Strawberries and bananas: · 5 crushed strawberries (31%) · ½ pressed apple · ½ mashed banana (24%) · ⅓ squeezed orange · 5 pressed red & white grapes · A dash of lime juice Fibre content: 3.3g Lemon, honey and ginger: 1½ pressed apples ¾ mashed banana ½ squeezed lemon Some squeezed orange Honey (1%) Ginger (0.2%) Fibre content: 2.5g Preparation

[93]The fruit composition of each of these smoothies is different but whichever smoothie a particular type of fruit is destined for, its preparation will be the same, as follows: Bananas : the whole banana minus its (inedible) skin is used. It is processed by being mashed. Soft fruits : these include blackberries, strawberries, raspberries etc. The whole fruit is used. They are processed by being crushed. They are then sieved but this is a large holed sieve designed to remove only the inedible twigs, leaves and stalk (or calyx). The seeds go through the sieve and are included in the smoothies. Oranges and other citrus fruits: the rind and pips are removed from fresh oranges which are then squeezed: the resultant juice contains all the orange cells as well as just the juice. (Each segment of an orange is comprised of tiny pockets visible to the eye made of flesh containing the juice and these pockets are known as orange cells). Most orange juices, unless freshly squeezed, do not contain the orange cells. Again the only parts of the orange unused by Innocent are those which are inedible. Apples and other hard fruit (eg pineapple): these are pressed (with skin on) on and then sieved. The purpose of the sieve is to remove the skin, pips and larger pieces of flesh. So the skin, which is edible in the case of an apple, is discarded. Edible chunks of flesh which don’t go through the sieve are also discarded.

[94]Mr Reed’s evidence was that Innocent aimed to balance its smoothies so that they comprised 60% whole crushed fruit and 40% fruit juice but accepted in many cases the balance was more 50/50. In any event, we do not agree with Mr Cordara that it is correct to describe the smoothies as a liquidised fruit salad. In the case of hard fruits, including apple, edible parts of the fruit are discarded and not used in the smoothie. None of the smoothies are equivalent to liquidised fruit salad: it would be more accurate to say that they are roughly 50% liquidised fruit salad and 50% fruit juice. Appearance

[95]The smoothies were sold in clear plastic 250ml bottles or in opaque 1 litre tetrapaks. The colour of the smoothies entirely depended on the fruit ingredients: for instance, the strawberries and bananas one was a pale pink, the mangoes and passion fruit one a strong orangey yellow, and the kiwis, apples and lime one a pleasant shade of green. They were all appetising to look at. Consistency

[96]We would describe the smoothies as having the consistency of a moderately thin soup, but being nevertheless easy to drink. When poured from a glass they leave a fibrous coating on the side of the glass. They could be drunk with a wide straw but this would not be so easy with a narrow straw. Although the fruit was entirely homogenised, nevertheless small seeds could be seen, such as strawberry seeds in the smoothies containing strawberries.

[97]We also found the smoothies to be quite thick compared to other drinks, except perhaps tomato juice to which it is comparable in thickness. We found that orange juice, even orange juice with extra bits, is much thinner and does not leave a thick residue on the glass.

[98]Mr Reed’s evidence was that Innocent chose to sell the 250ml bottles of smoothie, which are intended for immediate consumption from the bottle, with a 35 mm aperture because this makes it less messy to consume the thick product. Using the industry standard 28mm aperture could lead to the smoothie glugging and spluttering when consumed. He rejected HMRC’s suggestion that Innocent chose the large aperture to make the product look bigger: he said the reverse would be true. The laws of physics would require a taller bottle for a narrower aperture, and height (not width) equates with “large” in the eyes of a consumer. Ms Mitrophanous pointed out that some soft drinks have apertures wider than 28mm, and while we agree, we don’t think it detracts from the valid point made by Mr Reed, which we find to be that case, that due to its consistency a smoothie would be less easily drunk from a narrow necked than wide necked bottle. Nutritional value of smoothies

[99]It was Professor Strain’s evidence that smoothies are good for you (although not as good as the raw fruit itself- see paragraph 79 above) and we accept this evidence. 100.Professor Strain’s evidence was that smoothies are quite different to fruit juice because the latter basically contains water and simple sugars, whereas the former contain complex carbohydrate in the form of dietary fibre. Pectin is the predominate type of fibre found in fruit, although there are other soluble fibres found in fruit but they tend to have properties similar to pectin. Professor Strain’s evidence therefore concentrated on the properties of pectin, particularly as that was the only fibre in respect of which he was able to find reports, as mentioned below. 101.Pectin is a type of soluble dietary fibre. When consumed, its gelling properties (relied on by jam-makers) cause it to form gels in the inner intestine that slow down the absorption of sugars and fats and slow the transit of food in the inner intestine, causing the eater to feel full. Pectin is not really digested in the small intestine but reaches the large intestine where it is fermented by bacteria. In the lower intestine it also has the effect (opposite to the effect it has in the upper intestine) in that it speeds up the transit of food. 102.We understood Professor Strain to mean that fruit is good for you not only for the nutrients it contains but also because of the action of pectin which slows down and spreads out the body’s absorption of nutrients and provides the consumer with feeling of fullness which limits how much he wants to eat. In contrast, the sugars in fruit juice are absorbed immediately into the blood stream and provide instant energy and do not provide a feeling of fullness. 103.Professor Strain also mentioned, as is well known, that the government recommends that each person consumes five portions of fruit and vegetables each day. The Government allows a fruit juice to count towards 1 portion of that 5-a-day (but only one no matter how much fruit juice is drunk). Innocent fruit smoothies are permitted to claim 2 portions of the 5-a-day (although no more than 2 no matter how much is drunk). This is because they count as one portion of fruit juice (as they contain at least 150ml of fruit juice) and one portion of fruit (as they contain at least 80g of whole fruit). 104.It was Professor Strain’s evidence that many nutritionists disagree with the Government’s acceptance of fruit juice counting towards 1 of the 5-a-day because fruit juice almost entirely lacks the fibre present in fruit. He did not think that they would disagree with a smoothie counting towards 2 of the 5-a-day. We note, however, that Professor Strain had not appreciated until cross-examination that Innocent smoothies were a mixture of crushed fruit and fruit juice and were not therefore composed as he had thought of 100% of the edible parts of the fruit from which they were made. In conclusion, it seems to us that if fruit juice should not count towards the 5-a-day then an Innocent smoothie should only count towards 1 of the 5-a-day, discounting its claim that relies on the 150ml of fruit juice. How much fibre is in a smoothie?

[105]It was part of Innocent’s case that its smoothies were significantly higher in fibre than fruit juices. We therefore had evidence on the fibre content of the smoothies and some fruit juices. 106.The Tribunal found that the claims for fibre content on the labels of products may not be accurate. In the UK claims for fibre content can be based on theoretical grounds rather than actually proving that the fibre was there in the product. This was borne out by in relation to an orange juice which had “extra juicy bits” which was produced to the Tribunal as a comparator. The package shows fibre of 1.8g per 200mls. Innocent had it tested using the industry standard AOAC, however, and the result was that it had less than 0.1g fibre per 100mls. 107.Professor Strain’s evidence was that fruit juice would normally contain only a trace of fibre. An example of this was Innocent’s own orange juice which has 0.75g per 250mls. We do note Mr Reed’s evidence that the apple juice used by Innocent, which results merely from apples being pressed and sieved and no other treatment, will contain pectin, and indeed is the largest source of pectin in the smoothies. Pectin is a soluble fibre and is normally present in untreated apple juice. As Mr Reed explained, this is not true of orange juice as the pectin in oranges is in the inedible and discarded rind, so that even unfiltered orange juice will not contain pectin. 108.The fibre content of Innocent’s current smoothies is set out in paragraph 92 above with the ingredients and varies from 2.5g to 4.8g per 250ml bottle. Innocent also produced the fibre content as found by independent laboratory analysis in the past, although this was in respect of earlier smoothie recipes and therefore does not entirely correlate with their existing range. The results are however, roughly in line with their existing range showing as it does variation between fibre content of 2.25g -5.25g per 250ml. 109.A set of AOAC fibre measurements which included the pectin levels were provided to Professor Strain in February 2008 and these were (per 250ml smoothie): Total fibre as measured by AOAC (grams) Pectin (grams) Strawberry and banana 4 2.25 Oranges banana & pineapple 2.25 0.9 Mango and passion fruit 6 3.75 Pomegranate blueberry and acai 3.25 1.4 Blackberry and rosehip 3.7 2.25 Mango and carrot 3.25 2.3 Pineapple banana and coconut 2.5 2.2 Goji, mango and guava 5.75 3.45 Cranberry and raspberry 5.75 3.8 110.In conclusion, we find that almost all the Innocent fruit smoothies do contain at least 3g of fibre. Many do not contain 3g of pectin. Professor Strain’s evidence was that although the studies concentrated on pectin, the other fibres in fruit were also soluble dietary fibres and similar in action to pectin. It was therefore the total quantity of fibre and not just the quantity of pectin that he found significant. How much can be eaten at a single sitting? 111.As mentioned above, Innocent sells its fruit smoothies in two sizes. One is a plastic bottle which holds 250ml. It also sells a 1 litre Tetra Pak (which it sees as 4 250ml servings). Mr Reed’s evidence was that when Innocent launched on the market 12 years ago they intended to use a 330ml bottle as that was the industry standard for orange juice and other soft drinks. Market research carried out by Innocent then showed that this was too big and that 250ml was seen by consumers as the right sized portion. HMRC suggested on the contrary that the reason for the smaller bottles was in fact to reduce the price of what is an expensive product. Mr Reed disagreed that this was the reason for the 250ml bottle and we accept his evidence. 112.Professor Strain’s evidence addressed why smoothies were filling. He explained that the pectin and other soluble dietary fibre present in the smoothies leads to a feeling of fullness that a consumer would not get with fruit juice and limits how much smoothie can be drunk at a single sitting. 113.In Professor Strain’s view, it is likely that 3g or more of pectin and/or other soluble dietary fibre will produce a feeling of fullness and that as Innocent’s 250ml smoothie bottles contain at least 3g of soluble dietary fibre they will therefore lead to a feeling of fullness that you would not normally get with a beverage. 114.The evidence which Professor Strain relied on to substantiate his views was as follows: · A 1977 study [1] published in the Lancet which found that pureed apples were more satisfying that apple juice; · A 1997 study [2] of the effects of pectin which found that 5g of pectin increased satiety in much the same way as larger quantities. It did not test consumption of lesser quantities of pectin; · A 2007 study [3] found 2.8g (but not 1g) of an alginate-pectin drink decreased intake of food by participants. 115.Ms Mitrophanous, as mentioned above, criticised his conclusion on a number of grounds but mostly because she considered that the figure of 3g was insufficiently evidenced. In particular, the second study was of 5g of pectin or above, and the third study was of 2.8g of a mixture of salt and pectin, and, further was a study of obese women and only those with dietary restraint showed a response. His response was that the second report showed that there was no altered dose response above 5g (ie that there was the same response at 5g, 10g, 15g and 20g) suggesting that the cut off point was likely to be below 5g. The third report showed that with the salt/pectin mix (which he accepted might have more of an effect than pure pectin) there was no response at 1g but a response at 2.8g in dietary restrained obese women. He considered that it was reasonable to conclude from these reports that 3g or more of pectin would lead to a feeling of fullness in most people. He agreed that there was an element of speculation in his conclusion. 116.Professor Strain agreed that these reports look at pectin levels whereas Innocent’s fruit smoothies often have less than 3g of pectin. His point is that the smoothies contain other soluble dietary fibres as well as pectin and these act in much the same fashion as pectin and that therefore in his view it is the overall fruit fibre content that is important. 117.Ms Mitrophanous referred the Professor to an extract of another report [4] to which he had not referred. This indicated that having a breakfast of high fibre (12g) or of low fibre (3g) had little influence on how much was eaten at lunch. Professor Strain said he had been aware of this report but had not considered it to be relevant as it did not differentiate between soluble and insoluble fibres. Nevertheless, he pointed out that it supported his view that 3g of fibre was as effective as a higher quantity. 118.Ms Mitrophanous also put it to Prof Strain that his conclusion was unfounded because she did not understand how a person could be feeling full at 3g of fibre when the daily recommended intake of fibre is 24g. His reply was that at 3g you might not feel totally full but you would start to feel full. In any event, we note that the 24g refers to all fibre whereas his evidence was restricted to the soluble fibre found in fruit and we had no evidence on the effect of insoluble fibre. 119.We had anecdotal evidence that some people might drink well over 250ml in a single sitting (such as a person who reported on Innocent’s own website that he had drunk an entire tetrapak before reaching the checkout). On the other hand, the Incite research showed that 64% of respondents agreed with the statement that “smoothies fill me up”. The evidence of Professor Strain, which we accept, is that most people would feel uncomfortably full if they drank much more than 250ml of Innocent’s fruit smoothies in a single sitting. This is consistent with our own experience of sampling the products and consistent with the market research.

[120]HMRC challenged this evidence pointing out that people do eat food (such as breakfast) at the same time as consuming a smoothie. So it is clearly possible to have more than just 250ml of smoothie at any one sitting. While we agree with this, we find there is a distinction between smoothies and other food, even other food high in other fibre, in that smoothies contain soluble fibre. 121.We find that an average person could easily consume a 250ml bottle of smoothie in a single sitting and both members of the panel have done so. Our experience is that we would not choose to consume much more than 250ml in a single sitting as they are filling. We expect that most people would be unlikely to have much more than 250ml in a single sitting. We recognise that it is quite normal to have a 250ml smoothie with other food, but nevertheless we find most people would feel uncomfortably full if they had much more than 250ml of fruit smoothie in one go. Consumption of smoothies How are they consumed? 122.Innocent clearly intend the 250ml bottle of smoothie to be drunk from the bottle. Mr Reed’s evidence (which we accept) was that the wide aperture was designed to facilitate it being drunk from the bottle. The tetrapak has to be decanted but it is still intended to be consumed by being drunk. (As with anything, there are exceptions, and Mr Reed knew of one elderly person who liked to consume his smoothie with a spoon from a bowl but he recognised that this was very much the exception.) Innocent did not dispute that smoothies are drinks and intended to be drunk. Where are they sold? 123.We were informed that Innocent’s sales of the 250ml bottle break down as to roughly 50% sold in supermarkets, 30% in high street outlets such as Starbucks, Boots and Eat, and 20% in contract catering sites such as workplace canteens. In supermarkets they are sold in the chilled fruit drinks category and in the other places they are sold in chilled drinks cabinet. 124.Mr Reed said that in supermarkets Innocent would like their smoothies to be displayed in the fruit and vegetable aisles (as he says smoothies are displayed in this way in the US and because he sees them as an alternative to fruit) but so far supermarkets in the UK persist in almost invariably displaying them in the chilled drinks cabinet. 125.In the chilled drinks cabinets, smoothies are normally displayed in one section on their own at an end, beyond the most expensive fruit juices (juices from concentrate being furthest away from them). Ms Mitrophanous points out that the explanation could be that supermarkets just chose to display what they see as chilled drinks in increasing price range, running from juice from concentrate, juice not from concentrate, to freshly squeezed juices to smoothies. Mr Reed points out that supermarkets are limited as to where they can be displayed by the need to keep them chilled. Where are they consumed? 126.It was Mr Reed’s evidence which we accept (it is certainly consistent with our general knowledge) that few restaurants or pubs offer smoothies, whether with meals out in the evening or otherwise. On the other hand, they are clearly consumed in coffee bars as Mr Reed’s evidence is that Innocent first started selling their smoothies in the coffee bars of West London in 1999 and (from above paragraph 123) even now a significant percentage is sold to coffee bars. Nevertheless, it is our conclusion that only a minority of smoothies are consumed in a social or entertainment context. 127.From the evidence that 50% are sold in supermarkets, and from other market research showing that they are consumed at home, in the office or on the go, we conclude that there is no single predominant place in which they are consumed. Why do people drink Innocent smoothies? Pleasurable to drink 128.The panel found the sample smoothies to be delicious. HMRC describes the smoothies as tasty, pleasurable to drink, refreshing and delicious. Not surprisingly the Appellant agrees. From the spontaneous reviews on the Appellant’s website it is clear their customers agree too. 129.The Incite research showed that 61% of respondents agreed with the statement that smoothies were a treat. But, showing the inherently variable nature of market research, when asked for their reason for consumption, however, only 17% of respondents put “treat” as their reason, presumably having something else as their main reason for consumption. 130.HMRC’s point is (as per the Bioconcepts test) that people drink them because they give pleasure. 131.We agree with Mr Reed that food must taste good if people are going to eat it: that smoothies taste good does not tell us if they are consumed because they taste good. The market research suggests that the main reason for consumption in perhaps 1 in 5 cases is because they taste good. We find that they do taste very good, that unless they tasted good no doubt very few would be sold. Incite also found that roughly half of people said they bought them because they liked them rather than because they were healthy (see paragraph 134 below).

[132]Our conclusion is that they are pleasurable to drink and that this is the main reason for consumption in a significant minority of cases. Social occasions 133.Would they be offered to guests in the same way a tea, coffee or alcoholic drink might be offered? The Appellant’s counsel said it is not the custom of people in this country to offer a smoothie to a guest as a drink although Mr Reed’s evidence was that he himself would offer smoothies to a guest. We discounted this evidence somewhat as it is almost inevitable that the manufacturer of the product would be likely to offer it to guests. We consider that smoothies are pleasant and easy to drink. We find that smoothies, although relatively new to the UK market, are nevertheless fairly well known, and that if a host happened to have some smoothie in their fridge, it is the type of drink that could be offered to a guest in addition to the more everyday drinks such as tea and coffee. Heath reasons 134.As mentioned above, under Government labelling a single smoothie counts towards 2 of a person’s 5-a-day and this may be one of the reasons why they are consumed. The Kantar research shows that 60% of people who choose to drink a smoothie (these are all smoothies including yoghurt bases thickies) say they do so because it is healthy. This “score” of 60% is much higher than for other drinks. Water scores 39%, fruit juice 34%, milk 28% and the rest are much lower. However, the answers generated by market research depend much on how the question is asked and what alternatives are given. A different set of questions posed in Innocent’s commissioned market research ( Incite) gives a result of only 23.4% of consumers choosing it for health reasons. It also shows that 53% agreed with the statement that they bought smoothies because they liked them and not because they were healthy (we were not told how the remaining 47% was divided between the “disagrees” and “don’t knows”).

[135]The AC Nielsen Scantrack evidence shows that the biggest month for sales of smoothies is January. HMRC accepted Innocent’s interpretation that this happens because people are looking for healthy foods in January to compensate for the Christmas festivities. 136.We find that the products are perceived by consumers as good for their health (and the Professor’s evidence is that they are indeed right to think this). We find that seeking to improve their health is one of the criteria which lead people to consume them. But we are unable to conclude from this evidence that it is the main reason for their consumption. Fruit juice substitute?

[137]Mr Reed’s evidence was that Innocent had moved into the fruit juice market because of market research which showed that people would buy a smoothie and fruit juice at the same time: so far from being a substitute for fruit juice a smoothie was complimentary to it. HMRC, on the other hand, pointed to the fact that the Deeper Dive market research concluded that Innocent’s route to expansion “is pulling across” chilled juice buyers. 138.Our conclusion on this is that a fruit smoothie is not by any means an identical product to fruit juice and someone wanting to drink a fruit juice might not on all occasions chose to drink a smoothie in substitution – but that there might well be some occasions where such a person would see it as an alternative. Fruit substitute? 139.Innocent see their product as a substitute for fruit. Mr Reed’s evidence was that Innocent’s business started because he and his fellow directors perceived that there was a market for a product that made it easy to consume fruit. Innocent’s Incite research backs this up in that 72% see smoothies as an easy way of getting the fruit they need. As mentioned above, the Kantar research shows that 60% of consumers put health as their main motivation, which is comparable to the 70% who give this as the motivation for eating fruit. The Incite research also shows that only 29% of people “much prefer” eating fruit to a smoothie. 140.This evidence does not tell us that the customers would have consumed fruit had smoothies not been available: if anything it suggests that it was the consumers’ perception that their diet did not include enough fruit that led to a market for smoothies. We do not find that a smoothie is often bought instead of fruit: rather we find that someone buying a drink or a snack might choose a smoothie over the alternatives because it is made from fruit. Breakfast substitute? 141.Innocent gave evidence, which we accept, that a 250ml Innocent Smoothie contains in many cases more fibre than the average helping of other foods that might be consumed for breakfast. The fibre measurements of the Innocent smoothies were taken from independent food laboratory’s test: the fibre measurements of the comparators were taken off the packets, which we find from Professor Strain’s evidence is if anything likely to over- rather than under-state the fibre contents. 142.Research commissioned for Innocent shows that orange juice consumption spikes at breakfast. Smoothie consumption is also slightly higher at breakfast time but doesn’t spike. Instead it gradually trails off during the day.

[143]From this we are unable to conclude that a smoothie is consumed in substitution for breakfast: it might well be consumed in addition to other foods at breakfast or possibly instead of fruit juice. We accept that some people, especially those in a hurry, might consume nothing but a smoothie for breakfast. The fact it is easily drunk from a bottle makes it very suitable for consuming “on the go”. Snack? 144.Mr Reed’s evidence was that apart from breakfast alternatives, Innocent see snacks as the other main competitor for their product. In his view, people use smoothies as snack food to tide them over from one meal to another. We agree from the nutritional evidence given by Professor Strain that they are suitable to do this: because of their soluble fibre content they have a filling effect. Mr Reed produced evidence that an average Innocent smoothie contained as much fibre as a supermarket own-brand pot of snack fruit salad and on average about 3 times as much fibre as other snacks such as muffins, cereal bars, crisps, and yoghurt pots. 145.The Incite market research showed that 53% of respondents saw smoothies as more of a snack than a drink. We note that Ms Mitrophanous questioned this evidence on the basis that the 53% was calculated from 3 groups of respondents only one of which had respondents of which more than 50% considered it to be more a snack than a drink. But as Mr Reed pointed out the average would have been calculated on the basis of the numbers of respondents in each group and this is why the overall figure was 53%.

[146]Two results from qualitative pieces of market research, involving very small focus groups, Project Mint and Deep Dive, suggest that smoothies are seen as a food substitute. The Project Mint result was that the primary reason given for consuming smoothies was to fill a gap. Deep Dive suggested that people bought them as meal or snack replacements. There were similar results from the Tangible Shopper focus group. However these pieces of research were not, as Innocent accepts, statistically significant, although as Mr Reed points out businesses clearly value the insights from focus groups as they often act on them. 147.We conclude from the evidence of how the smoothies are produced and the evidence of their nutritional value that Innocent’s fruit smoothies are apt to be used as a snack. Indeed, the evidence from market research that around 50% of consumers regard it primarily as a snack rather than as a drink is entirely rational. We bear in mind, though, that the smoothies are also 50% equivalent to fruit juice and that around 50% of consumers did not consider them primarily as a snack rather than a drink. Thirst quenching properties 148.Fruit smoothies have a high water content of 84% which is not that much lower than fruit juice at 88%.

[149]We accept Innocent’s point that water content does not necessarily indicate whether a drink or food is suitable to quench thirst. Their Incite market research indicates that only some 13% of purchases of the product were made in order to quench thirst. The Kantar market research indicates that only some 6% of smoothies (all smoothies including thickies) were bought to quench thirst. Ms Mitrophanous thinks that this low figure in part may be explained in that consumers consider smoothies too expensive to drink merely to quench thirst. Water would do the job at much lower cost. 150.Innocent produced a graph based on 2009/10 figures from AC Nielsen Scantrack which Innocent considered showed that their sales were lowest in summer: we accept the graph shows that their highest sales are in January (we have referred to this above). However we agree with HMRC that it seems to show a small (but lower) peak in June and July with lower sales in autumn and the lowest sales in December. Colas and bottled water show a much greater peak in the summer. This evidence suggests to us, consistent with our tasting of the product, that the products would not be perceived to be as thirst quenching as thinner drinks such as juice and water and soft drinks. 151.Part of the Appellant’s case was that smoothies could not in fact quench thirst. The evidence on which both experts agreed was that most people would be at least 1% dehydrated before they felt thirsty. Professor Strain’s evidence was that the 1% was conservative and that most people would lose 2% body weight through water loss before feeling thirsty and we accept that a person will not feel thirsty until they are 1-2% dehydrated. 152.Professor Strain calculated that if the average person weighed 70kg (likely to be an underestimate) then to re-hydrate that person would need to consume 700ml of water to replace the loss of 1% of body weight through dehydration. As a smoothie is only 84% water, this means he would have to drink 833ml of smoothie to rehydrate. The average person would probably have to drink more. But in the Professor’s view a normal person could not drink 833ml of smoothies in a single sitting: most people would choose not to drink more than 250ml. They would feel uncomfortably full long before they reached full hydration. This is not true of fruit juice which is much easier to consume in large quantities. 153.However, he accepted that (as a person does not feel thirsty until about 1-2% dehydrated) a person might cease to feel thirsty before they reach full hydration. He also accepted that if a person was fully hydrated, sipping smoothies throughout the day might keep a person fully hydrated (we note there was no evidence that smoothies were consumed in this manner). 154.We accept his evidence that it would be difficult to rehydrate fully from drinking a smoothie. But we also think it unlikely a thirsty person would drink 1 litre of water in a single sitting. In any event, the question of whether a drink can rehydrate a person is a different question to whether people drink smoothies because they are thirsty. 155.Miss Gearey mentioned a study carried out by Campden BRI in 1996 [5] to test the thirst quenching properties of drinks. The study’s conclusions were that a drink with acidity and/or astringency and/or fruitiness and/or strong flavour would be seen as thirst quenching by most people. Drinks with carbonation, sweetness and thickness would not be seen as thirst quenching. In her view smoothies would be perceived as thirst quenching just not as thirst quenching as some other drinks such as fruit juice and water.

[156]Our conclusion is that a smoothie has some capacity to quench thirst (even though we agree they cannot be used to fully hydrate a dehydrated person) but that quenching thirst is not one of the main reasons why they are drunk. Are they consumed for their fortifying effect?

[157]Smoothies are clearly not fortifying in the alcoholic sense: they contain no alcohol. We have said above (see paragraph 47) that we do not think “fortifying” in the Bioconcepts test means fortifying in the sense that all foods are good for you and fortify the body. We accept Professor Strain’s evidence that the smoothies are less likely to provide a sugar rush than fruit juices or soft drinks because the soluble fibre forms gels in the stomach which slow the absorption of the sugars by the body. We do not think that Innocent’s smoothies are fortifying in the sense meant by the Bioconcepts test. Comparators Fruit Juice 158.Fruit Juice is a beverage: Exception 4 in Group 1 VATA says “other beverages (including fruit juices and bottled waters)”. Even without this definition fruit juice is something which we find is sold and drunk as a beverage and we do not understand either party to dispute this. We also note that Warren J in Kalron considered fruit juice to be a beverage (paragraph 10 of his decision). 159.We were asked to consider as a comparator an orange juice with extra juicy bits. As reported above this was found to have only trace fibre and the evidence was that all fruit juice (with the exception of the unfiltered cloudy apple juice of the type used by Innocent in its smoothies) would only have trace fibre content. This is in contrast to Innocent’s fruit smoothies which have around 3g of fibre in each 250ml serving.

[160]We find that Innocent’s fruit smoothies contain significantly more fibre than most fruit juices. We find that fruit juices do not fill you up in the way that fruit smoothies do and fruit juices are more apt to quench thirst. As noted at paragraph 158 above, fruit juices are defined as beverages in the legislation. Nevertheless we find their fibre content cannot be decisive: as we have said as a matter of law the effect on the digestive system is not the test (see paragraph 58) and in any event on the facts it would seem at least that unfiltered cloudy apple juice has more than a trace amount of fibre (and indeed is one of the main sources of fibre in the smoothies) yet all apple juice is standard rated as a beverage as it is a fruit juice. Alcohol and coffee 161.We put little weight on the fact that we have found that not much more than 250ml of smoothie would ordinarily be consumed in a single sitting. This is clearly true (although for other reasons) for many other drinks that are clearly beverages such as strong alcohol and espresso coffee. 162.We also note that strong alcohol dehydrates and that tea and coffee are diuretics. They are not ideally suited to re-hydrate or quench thirst. The fact that smoothies are not particularly apt to quench thirst does not prevent them being beverages. Alcoholic drinks, tea and coffee are beverages. Soft drinks 163.We consider it a matter of general knowledge that smoothies (made of fruit) are healthier for people than soft drinks (which typically contain added sugar and/or artificial sweeteners and other additives). Soft drinks are standard rated beverages. Other smoothies 164.The High Court in Kalron (albeit on different evidence) found smoothies made from nothing but the whole edible parts of fruit and vegetables to be beverages. 165.Innocent consider (and HMRC agree) that their children’s smoothies are standard rated beverages. HMRC’s view is that this indicates that the adult smoothies (in issue in this case) are similarly standard rated beverages. Innocent point out that the products are different: the children’s smoothes are 80% fruit juice to 20% crushed fruit. They see them as being bought as drinks and not fruit substitutes. 166.Ms Mitrophanous points out that the children’s smoothies appear to have similar levels of fibre to the adult smoothies. Mr Reed explained this is because the main fruit juice is unfiltered cloudy apple juice produced as explained in paragraph 107 and high in pectin. Ms Mitrophanous also pointed out that they were marketed in the same way and in the same supermarket chiller cabinets as adult smoothies. 167.We cannot derive any assistance from Innocent’s (or indeed HMRC’s) view of the law and in particular whether they are right to regard Innocent children’s smoothies as beverages within the meaning of Group 1. What we do consider to be of assistance is the fact that it shows that apple juice (which is standard rated) can (if unfiltered cloudy apple juice) have similar pectin levels to smoothies and it follows from Professor Strain’s evidence must have the same “filling” effect. This indicates to us that the pectin level and the filling effect are not decisive as to whether something is a beverage within the meaning of the law. Milk 168.Neither Counsel seemed to consider that useful comparisons could be drawn from milk. Milk is an item overriding the exception from zero rating that applies to beverages which would suggest that the legislators considered it to be a beverage (or at least that it might be considered a beverage). It was therefore expressly excluded from standard rating as a beverage as it was Government policy that milk should not attract VAT. Nevertheless, the Tribunal in Alpro decided that fruit flavoured soya milk was not a beverage indicating that milk itself is not a beverage. Mr Warren in Kalron indicated that he thought that milk is not ordinarily a beverage although he reached no firm conclusion on this (paragraph 68 of his decision). 169.Milk is comparable to smoothies in that both could be seen as snacks in that milk is high in protein and smoothies are high in fibre. They are not a liquid containing little else but water, sugar and flavour such as is the case with (say) many soft drinks and fruit juice. On the other hand milk being protein based and smoothies being based on fruit and fruit juice are two very different things and we are wary of making a direct comparison. Further milk is commonly cooked with and when cooked, is not consumed as a drink. It is also (to state common knowledge) frequently eaten with a spoon out of a bowl by the presumably significant proportion of the population who eat breakfast cereals. Although we heard no evidence on this, we consider it is possible (at least amongst adults) that milk is eaten at least as much as it is drunk. Smoothies do not have this dual identity as drink and cooking and breakfast ingredient. Like others before us, we regard milk as being in a special category. 170.When it is eaten rather than drunk, milk is clearly consumed other than as a beverage. Nevertheless the draftsman considered that milk, a nutritious liquid high in protein, could be viewed as a beverage. If he had not, he would not have made it an overriding item. This suggests to us that when it is drunk as a drink it is being consumed as a beverage (but we are not called on to decide this). We are unable to conclude from this that the fact that both milk and smoothies are more nutritious than most beverages means that a smoothie is not a beverage. If anything, these considerations drive us to the opposite conclusion. Everyday language 171.We have commented above in paragraph 37 that “beverage” is not a word in common use. Nevertheless, we were shown during the hearing a number of places where it has been used generally and used in the context of a smoothie being a beverage. The author who wrote the internal presentation for Innocent showing the results of the various market research reports thought of smoothies as a beverage because they included strap lines such as “the primary motivations for consuming most beverages are ‘quench’ and ‘perk’. The only exception is smoothies where consumption is primarily motivated by ‘pure health’, ‘treat’ and ‘fill a gap’” and “the majority of smoothies consumption is for health reasons, in contrast with the rest of the beverage category” and lastly “Smoothies are unique among beverages….”. But this is (presumably) the view of a single person. We were also shown that a journalist had reported Mr Reed referring to smoothies as “beverages” in an interview. Innocent have won food awards – they won the title of best UK soft drink one year. They also enter the Quality Food Awards, a body which puts them in the category with “beverages” and not in the deserts and puddings or snack categories. And as HMRC noted, Professor Strain referred to a smoothie as a beverage in the Ocean Grown case although of course that was in the context of a case on the meaning of a beverage. 172.We are wary of drawing any firm conclusions on the case from this evidence as it is based on the language used by a very small number of people, although we note that it is all consistent with smoothies being beverages. We do consider that if the word “beverage” was to come back into common use that a smoothie could be listed as a “beverage” by the ordinary person. Conclusions 173.Innocent’s case in a nutshell seemed to us to be that its smoothies were the same as a fruit salad and that therefore they should be zero rated as fruit salad is zero rated; that fruit smoothies were not beverages under the Bioconcepts test; and that they are drinkable liquids and not beverages. 174.Our conclusions on the facts are: Smoothies are a delicious drink of pleasing appearance; They are thicker than most beverages but despite the thickness they are easily drunk (except perhaps with a narrow straw); They are good for you because they are made of fruit; They are filling due to their fibre content. Unusually for drinks, they require digestion. Their fibre content means that they can act as a snack and also means that most people would be unlikely to drink much more than a 250ml serving in a single sitting; They are intended to be and are consumed by being drunk; They are sold as drinks and they are displayed for sale in shops with other drinks. They are not usually sold with snack food; There is not one single overriding reason for consumption. We treat the market research evidence with some caution as explained in paragraph 86. We find that the main reasons for consumption are that they taste good, they are a treat, they are considered to be healthy, they are filling and they are seen as a snack; Innocent see their product as fruit substitute but we cannot agree. We find that there is little evidence that people who buy smoothies would have bought fruit if smoothies were not available and indeed Innocent’s own perception of its product’s appeal is that there is a market for a product which makes fruit easy to consume because people don’t eat enough fruit. We also find that fruit smoothies are to some extent in competition with fruit juices as they are sold alongside them and Innocent’s own market advisers see their product as being in competition with fruit juice; and There is no one main place or time of consumption of fruit smoothies: they are drunk with breakfast, on the go, in the office, in a coffee bar or cafeteria and so on. They may be consumed with food, for pleasure or instead of a snack or another drink. 175.We are unable to agree with Mr Cordara that Innocent’s fruit smoothies are the same as a fruit salad and should have the same VAT treatment. On the facts we have found that the composition of Innocent’s fruit smoothies is on average 50% fruit juice and 50% liquidised fruit salad. On the law, Warren J has ruled that liquefaction might be (and indeed was in the case of Kalron ) sufficient to convert a fruit salad into a beverage. We consider whether liquefaction has had a similar effect on Innocent’s fruit smoothies. 176.Innocent puts the case that their smoothies are a drinkable liquid rather than a beverage. Are their smoothies like a liquid medicine or dietary supplement that is consumed by being drunk but is not a beverage? As we have said in paragraph 70, we think this turns on whether they are primarily consumed as a drink or beverage or for some other reason. If Innocent’s smoothies are primarily consumed as snack food this may mean that they are a drinkable liquid and not a beverage. 177.But we do not find that Innocent has made out its case on this. We bear in mind the market research which showed that slightly over 50% of respondents saw smoothies more as a snack than a drink. As we have said we are wary of putting too much weight on market research which gives different answers to similar questions from one survey to another and where answers depend so much on how the question was asked and what alternatives were given. We bear in mind that the smoothies are sold and marketed as a drink and with other drinks and are seen to some extent to be in competition with fruit juice. We find that the purpose of Innocent’s fruit smoothies is to be a drink and that there is no evidence they would be consumed if they were not in liquid form. 178.Our conclusion is that fruit smoothies are not merely drinkable liquids. They are drinks. But we recognise (as Warren J himself said) that there may be a category of drinks which are not beverages. 179.So we come full circle and consider whether Innocent’s fruit smoothies are beverages. We consider the Bioconcepts definition on which the Appellant places great weight. We do agree with Innocent that fruit smoothies are not apt to rehydrate and that quenching thirst is not the one of the main reasons for their consumption. We also agree they are not fortifying, which we think is limited in meaning to alcoholic and stimulating drinks. We think that they are very pleasant to drink and are drunk for pleasure but this is probably not the principle reason for consumption in most cases. 180.Our conclusion is that even under the Bioconcepts test smoothies display some characteristics of beverages in that a minority of people drink them to quench thirst or principally for pleasure. But we bear in mind that Bioconcepts is not an exhaustive test and there may well be beverages outside the Bioconcepts test. 181.We think we should consider the smoothies in a social context as Ms Mitrophanous suggested: is it the sort of drink which could be offered to a guest at the same time as or instead of drinks which are beverages? We find it is a pleasant drink that is drunk in some social situations (eg it is sold in coffee bars), that it is pleasant to look at and easy to drink. We think that it could be offered to guests as alternatives to beverages such as tea, fruit juice, alcohol, even though we accept Mr Cordara’s point that it is not typically offered to guests. 182.We consider the comparators such as fruit juice, alcohol, coffee, soft drinks, other smoothies and milk. We find that beverages can be small (such as an espresso) or large (a 500ml bottle of soft drink). They can be alcoholic or non-alcoholic. They can be a stimulant or not. They can be thirst quenching or not. They can be healthy or not so healthy. Many will contain little or no nutrition and most will not require digestion. Our conclusion is that there is a great deal of variation in beverages. As drinks, fruit smoothies are unusual (but not unique) in that they are quite thick and do require digestion: we do not think this is sufficient to mean that they are not a beverage if they are drunk as other beverages would be drunk. 183.We consider the ordinary meaning of the word beverage. We have some evidence that the word beverage is used to describe a smoothie and we think that if a list of beverages were to be produced people would not be surprised or think it an odd use of language to put Innocent’s fruit smoothies on such a list. We think this because it is drunk, it is intended as a drink, it is sold as a drink and it is very pleasant to drink. We think it would be socially acceptable to consume a smoothie in place of undoubted beverages such as tea or coffee or fruit juice. We recognise that it does have some snack-like properties and that it is seen as a snack by many consumers but we find that nevertheless it drunk as a beverage. The fact that it is also a food does not take it out of the category of beverages. If there is a category of drinks which are not beverages, we are satisfied that fruit smoothies are not in it. 184.Our finding is that Innocent’s fruit smoothies are beverages for the purpose of Group 1 of Schedule 8 VATA 1994. 185.This document contains full findings of fact and reasons for the decision. Any party dissatisfied with this decision has a right to apply for permission to appeal against it pursuant to Rule 39 of the Tribunal Procedure (First-tier Tribunal) (Tax Chamber) Rules 2009. The application must be received by this Tribunal not later than 56 days after this decision is sent to that party. The parties are referred to “Guidance to accompany a Decision from the First-tier Tribunal (Tax Chamber)” which accompanies and forms part of this decision notice. Barbara Mosedale TRIBUNAL JUDGE RELEASE DATE: 25 October 2010 [1] Depletion and disruption of dietary fibre: effects on satiety, plasma-glucose, and serum-insulin by Haber, Heaton and Murphy, The Lancet 1 October 1977 [2] Professor Strain had read the full version of this article but only an extract was available to the Tribunal. The full version was Effect of pectin on satiety in healthy US Army adults , by Tiwary, Ward and Jackson and published in the Journal of the American College of Nutrition 1997. [3] Novel calcium-gelled, alginate-pectin beverage reduced energy intake in nondieting overweight and obese women: interactions with dietary restraint status , by Pelkman, Navia, Miller and Pohle, published in The American Journal of Clinical Nutrition in 2007 [4] The Effect of high and low-fibre breakfasts on hunger, satiety and food intake in a subsequent meal by Burley, Leeds and Blundell in a 1987 edition of the International Journal on Obesity. [5] The Sensory Assessment of the thirst-quenching characteristics of drinks by McEwan and Colwill and published in Food Quality and Preference Vol 7 pp 101-111 (1996)

Innocent Ltd v Revenue & Customs [2010] UKFTT 516 (TC)

Cited in 9 later judgments