19389
ZERO-RATING – food – sports nutrition bar – an item of sweetened prepared food normally eaten with the fingers - whether 'cake' for the purposes of the exception – no – whether sufficiently similar to a traditional flapjack to be classified as cake – no – appeal dismissed – VATA 1994 S 30 and Sch 8 Group 1 Excepted item 2
LONDON TRIBUNAL CENTRE
TORQ LIMITED
Appellant
- and -
THE COMMISSIONERS OF CUSTOMS AND EXCISE
Respondents
Tribunal: Malcolm Gammie Q.C. (Chairman)
Tym Marsh
Sitting in public in London on 18th November 2004
Richard Mills of Mills & Co for the Appellant
Zoe Taylor of Counsel, instructed by the Solicitor of Customs and Excise, for the Commissioners
© CROWN COPYRIGHT 2005
DECISION
Introduction
- Torq Limited ("the Appellant") appeals against a decision of the Commissioners of Customs and Excise ("the Commissioners") that sales of the Appellant's product, known as "Torq Bars", are subject to the standard rate of Value Added Tax. The Appellant says that Torq Bars should be zero-rated.
The relevant legislation
- Section 30 of the Value Added Tax Act 1994 provides that a supply of goods or services is zero-rated if the goods or services are of a description for the time being specified in Schedule 8. Group 1 of Schedule 8 is headed "Food" and provides that the supply of food of a kind used for human consumption shall be zero-rated unless it is a supply in the course of catering or a supply of anything comprised in any of the excepted items, subject to a supply of certain items overriding the excepted items.
- We are not concerned in this appeal with catering or with any of the items overriding the excepted items. The appeal relates to item 2 of the excepted items, namely—
"Confectionery, not including cakes or biscuits other than biscuits wholly or partly covered with chocolate or some product similar in taste and appearance."
- Note 5 to Group 1 provides that—
"... for the purposes of item 2 of the excepted items "confectionery" includes chocolates, sweets and biscuits; drained, glacé or crystallised fruits; and any item of sweetened prepared food which is normally eaten with the fingers."
The issue for determination
- The Appellant concedes that a Torq Bar is an item of sweetened prepared food which is normally eaten with the fingers. It therefore accepts that it is "confectionery" for the purposes of item 2 of the excepted items. Briefly, however, the Appellant says that a Torq Bar is a cake because the Commissioners accept that "flapjacks" are cake and if a flapjack is a cake, then so is a Torq Bar. A Torq Bar is a low-fat flapjack and therefore a cake. Accordingly, the issue for our determination is whether a Torq Bar is a "cake" so as to fall outside item 2 of the excepted items.
The evidence
- We heard evidence from Nicola Parsons, one of the Commissioners' officers who had conducted the reviews leading to their decision that sales of Torq Bars should be standard-rated. Matthew Hart, the managing director of Torq Ltd, also gave evidence. He holds a sports science degree and worked as a fitness instructor before deciding to start his own business as a personal coach and personal fitness instructor. He was responsible for the development of the Torq Bar.
- The Appellant produced a bundle of documents, including articles reviewing different sports bar products and recipes for flapjacks. We also had before us the correspondence between the parties in which they had exchanged their respective views as to how Torq Bars should be categorised for these purposes. We were supplied with and sampled a variety of products, including the Torq Bar and its competitor products and a number of other cake, cake bar and cereal bar products.
- Finally, the Appellant showed us a video designed to illustrate how the "ordinary man in the street" – or, more accurately, the "ordinary cyclist at a cycling event" might view a Torq Bar. It was taken at 2 separate cycling events – the Beast of Brecon on 12 September 2004 and the Gorrick in Southern England on 24 October 2004. The video provided a few moments of light relief to everyone at the hearing (and to the distraction of those in a neighbouring Court). We do not think, however, that it added significantly to the other evidence that we received or assisted materially in resolving the issue for our determination.
The facts
- The Appellant's business is described as "sale of sports nutrition bars". Mr Hart told us that he decided to develop his own brand of energy bars because as an athlete he had been unable to find a product which met his nutritional requirements, was relatively easy to consume during exercise and tasted reasonably pleasant. He assumed that other athletes had encountered similar problems and that therefore there would be demand for a product that met these needs.
- He identified the ingredients that the product should contain to meet the nutritional standards he required and then worked over a period of about two years with a food manufacturing company, FM Foods Ltd, to develop what became the Torq Bar. The whole philosophy and design of the Torq Bar is based on its nutritional value. Mr Hart said that his aim had been to produce a moist, chewy energy bar. The problem that he found with other energy bar products was that they were hard and therefore difficult to eat at lower temperatures. He designed Torq Bars as a natural product without fat or chocolate. The idea is to layer the intake of carbohydrate whilst maintaining a very low fat content. Mr Hart explained that the human body absorbs different types of carbohydrate at different speeds. By designing the Torq Bar with different layers of carbohydrate the body is able to benefit from a natural and reliable release of carbohydrate into the bloodstream over the duration of the exercise.
- The Torq Bars come in 2 flavours – "Tangy Apricot" and "Sun-dried Banana". The bars comprise the following ingredients:
22.3% Oats
19.3% Raisins
18.5% Maltodextrin
14.9% Fructose Syrup
12.4% Apricot or banana
7.4% Puffed Rice
Water
0.77% D-Ribose, and
a vitamin and mineral mix
- Mr Hart said that fat was useless for athletes because it slowed the digestion. Maltodextrin had all the attributes of fat without being fat. It gave a fat texture to the product on the tongue and enhanced the flavour of the other ingredients. It was a fat simulator. The body is able to absorb carbohydrate from Maltodextrin very quickly, followed by the rice, followed by the oats. The fructose syrup (including the raisins) kicks in last of all. D-Ribose is a naturally occurring sugar.
- Mr Hart had been advised by FM Foods that the manufacturing process for Torq Bars was very similar to that of other flapjack products made by them. It was baked as a flapjack, cut and left to cool on a tray. He had been to the factory and seen for himself the manufacturing process, which was described in the documents as follows:
1. Mix dry mix in large bowl mixer for 30 seconds
2. Make a syrup from the meltodextrin and the water
3. Heat this syrup in a pan with the fructose syrup
4. Add this hot syrup to the fruit in the bowl chopper and process for 1 minute
5. Add this puree the dry mix from stage 1 and mix for 3 minutes, scrape the sides then mix for another 15 seconds
6. Pin out onto trays
7. Make for 16 minutes at 250C
8. Mark at 112.5mm v 40mm
- Mr Hart said that the consistent response of those who had been approached and asked (as illustrated by the video evidence) was that people bought the product for nutritional reasons – to gain energy. The Appellant's current marketing strategy is to dominate the energy bar market for cyclists. Currently the main outlets for Torq Bars are cycling shops, the Torq website and direct sales at cycling shows and events. The Appellant's longer-term aim would be to expand into other sports. He anticipated that the product would become more generally known once the Appellant moved outside the more limited cycling market and began selling Torq Bars to organisations such as Premier League football clubs. He said that he could also envisage certain High Street outlets for Torq Bars, such as health food shops.
- Miss Taylor for the Commissioners suggested to Mr Hart that the ingredients of a Torq Bar were not the traditional contents of a flapjack or cake and that Torq Bars were not found in a usual 'cake' context: athletes did not consume cake in the course of their activities. Mr Hart did not accept this. He noted that Torq Bars contained at least two of the usual ingredients of flapjacks – oats and sugar – and said that there was no reason why the energy characteristics of Torq Bars could not be replicated in a cake. He agreed that Torq Bars were described as energy bars and never as cake and were not marketed as cake or flapjack. He agreed that they could not be described as biscuits but thought that they could be described as cake. In a cycling or athletic context they would be eaten with a cup of tea, both before and after exercise. Torq Bars were nutritional bars that aided recovery from exercise as well as providing energy during exercise. He agreed that a barrister might not ordinarily choose to eat a Torq Bar with an afternoon cup of tea but suggested that it might be beneficial if he or she did.
- The Torq Bars' main competitors are the High 5 Energy Bar and Science in Sport GO-Bar, Powerbar Performance Bars and Maxim Energy Bars. Mr Hart said that the SiS GO-Bar and the High 5 Energy bar were zero-rated. In this respect we saw invoices showing that those products had been supplied at the zero-rate. Miss Taylor on behalf of the Commissioners objected to this evidence on the basis that it was not evidence of the VAT classification of these products. We accept that the invoices do not mean that these products were correctly supplied at that rate, save that the SiS GO-Bar has been subject to a previous decision of this Tribunal, to which we shall come.
- The High 5 Energy Bar usually retails at around 99p per bar and the SiS GO-Bar at £1.04, whereas the Torq Bar retails at £1.15 per bar. Mr Hart said that the difference was entirely explained by VAT and that the difference in VAT rating meant that the Appellant currently made very little profit from the sale of the bars. He said that this was a significant impediment as compared with the competition.
- The Appellant produced a comparative assessment of different energy bars that had been published on the website of Sigma Sport. The products are introduced as:
"Concentrated food in a bar: no fuss, no mess, just a great tasting mix of carbohydrate, protein and fats in a pretty wrapper."
The website article compares the High 5 EnergyBar, Powerbar Performance Bar, Maxim EnergyBars, Science in Sport GO-bars and Torqfitness TorqBar. The size and nutritional value of each in terms of fat, sugar, protein and fibre content are listed for comparison.
- The article makes brief mention of their characteristics. The High 5 Energy Bar is evidently soft and easily palatable "and a lot less mess than jam sandwiches". The contents of a High 5 Energy Bar are listed as mixed dried fruits (pineapple, raisins, apple flakes, banana flakes), maltodextrin, puffed rice, lemon juice (with preservative: sulphur dioxide), vegetable glycerine, flavouring. The wrapper proclaims that it is 98% fat free.
- SiS GO-Bars are described as having much more texture than most of the other energy bars, "so that you actually feel that you are eating something rather than merely absorbing it". It is "sometimes a little too chewy to be eaten on the move but [is] definitely a great bar for when you've stopped for a break." The listed ingredients of the SiS GO-Bar are grape juice, maltodextrin, banana, soya protein, jumbo oats, dates, pineapple, crisp rice, puffed oats, apricot, lecithin, calcium, lactate, natural flavouring, malic acid, vitamins and minerals. The wrapper proclaims that it is over 98% fat free and with no added sugar.
- The Appellant also produced an article from Cycling Plus magazine for March 2004 in which sports foods and drinks were reviewed. The review included the High 5 Energy Bar, the Torq Bar, SiS GO-bar, Mars Bar, Jordan's Frusli Bar, Tradecraft GeoBar, Jacob's fig rolls and Jordan's Organic Cereal Bars. Flapjacks do not feature in the review.
- The article describes energy bars as being high in carbohydrates, rich in minerals and vitamins and low in fat. According to the article what all the products have in common are the main ingredients: glucose, sucrose and maltodextrin. It explains that—
"Maltodextrins are produced from starch and are chains of single glucose monomers formed into a linkage of glucose polymers. As a result maltodextrin consists of fewer large chains of glucose molecules compared to the same concentration of glucose monomers. Consequently, the osmolality (amount of dissolved pieces in a solution) of a maltodextrin solution is lower in comparison to a glucose solution. Because of this lower osmolality, maltodextrin solutions lead to better uptake from the stomach into the blood stream, thereby enhancing delivery to the muscles during exercise and enhancing your glucose restoration post-exercise. Maltodextrins are also less sweet than sucrose or glucose; the lower sweetness levels allow a higher concentration of carbohydrate to be added to technical foods without making the product unbearably sweet."
- The article describes the High 5 Energy Bar in the following terms:
"A good sized low-fat cereal bar (98% fat free) that offers a generous amount of fuel in the form of maltodextrin. With 58% high carbohydrate fruit and 22% cereal, the oaty texture was soft with plenty of fruit throughout. A definite fruity smell and flavour, though whether it resembled 'berry' is questionable. Available in a wide range of flavours that will appeal to all, as well as the sports bars and recovery bars which provide a generous 13g of protein."
- The SiS GO-Bar is described in the following terms:
"Coming in a range of great flavours, these cereal bars not only provide a high source of carbohydrates but also a good amount of protein making them an ideal snack before, during and after your workouts. With natural ingredients, high content of fruit (68% apple and 7% blackcurrant), no added sugar and added vitamins, this bar gives you the full package. Not only are they tasty, visually appealing and soft and chewy in texture, they are also 98% fat free."
- The Torq Bar is described as follows:
"Stylishly packaged, this bar is moist, chewy, easy to eat and extremely tasty; due to the adequate quantity of fruit (12% of this is banana), the wholesome and natural flavours would be appealing to many. Containing ribose (0.77%), that has been shown to increase the manufacture of ATP within skeletal muscle, the TORQ cereal bar is ideal for a high energy snack, on or off the bike. Low in fat and high in carbohydrate, and with a full range of vitamins and minerals, this bar is the softest of them all."
- Dealing with other products not recognised as specially designed high energy sports bars the article describes a Jordan's Frusli Bar as follows:
"For a bar not marketed as a sports bar (with the associated premium price tag) this bar is just about perfect. The impressive ingredients include conservation grade (almost organic standard oats), malt syrup (a source of sugar with a sprinkling of B vitamins) and no hydrogenated fats."
- It describes the Tradecraft Geobar in these terms:
"Another excellent cereal bar based on riceflour, barleyflakes, wheatflakes and oats but without hydrogenated fats. This is a high carbohydrate, easy to digest bar which is good for you ..."
- Jordan's cereal bars are described as follows:
"Excellent cereal bar containing organic ingredients. The nuts, oil and sunflower seeds contribute more fat than the frusli bar. The oils are digested and absorbed more slowly compared with sugars and so contribute to a feeling of fullness. A couple of these bars would be useful mini meal during a day of cycle touring."
- The article compares the calories provided by each product and which ingredients provide them. The non-sport bars provide a higher number of calories from fat, as you might expect. Mr Hart agreed with what appeared from this article that one possible categorisation of Torq Bars and its competitors were as cereal bars.
- Miss Parsons gave evidence as to her reasons for concluding that Torq Bars were standard-rated. As such she confirmed the content of the Commissioners' letters. Miss Taylor also asked Miss Parsons what she regarded as the ordinary ingredients and characteristics of a cake. Her description did not to our mind explain why flapjacks are cake and if they are cake, then her description cannot have been comprehensive of what can be regarded as cake. Asked why she did not regard Torq Bars as flapjacks, Miss Parsons said that she had referred to the Commissioners guidance on the subject and had concluded that Torq Bars were not the same as a traditional flapjack. Torq Bars did not have traditional flapjack ingredients and involved different cooking processes. She said that they were larger in size than ordinary flapjacks and were more moist than flapjack. She said they looked more like cereal bars.
- Mr Mills for the Appellant pressed Miss Parsons to explain the distinction that the Commissioners were seeking to draw between cake, flapjacks, cereal bars and the different forms of energy bar. Miss Parsons was not able to shed much light on the matter. This is perhaps unsurprising. What the Commissioners are seeking to do in their categorisation of different products is to make sense of legislation, the application of which constantly has to be adapted as new products develop, where the underlying policy of the legislation can perhaps most charitably be described as opaque. What was apparent was that she had considered Torq Bars as an individual product and asked whether, in line with the Commissioners existing guidance, they were a product that could properly be described as a cake. She was not prepared to be drawn more deeply into matters of comparison with other products but she confirmed that consideration of the social reasons underlying the standard or zero-rating of particular foods had not entered into her decision.
The Commissioners' Guidance
- At this point, we should set out the Commissioners' published guidance, so far as relevant to this case. Notice 701/14 explains when food can be zero-rated. Paragraph 3.4 states that most traditional bakery products, such as bread, biscuits and cakes are zero-rated but that some confectionery items are standard rated. The summary indicates that flapjacks are zero-rated but cereal, muesli and similar bars with honey or other added sweetening matter are standard-rated.
- Section 4.6 of the Notice is headed "Sports products" and paragraph 4.6.3 is as follows:
"Cereal/Fruit Bars: Standard-rated items include compressed fruit bars. Consisting mainly of fruit and nuts, with added sweetening matter (including honey) and also cereal bars, whether or not coated with chocolate, with the exception of bars which qualify as cakes – see paragraph 3.6 [which lists examples of zero or standard-rated confectionery]. Standard-rating applies to any product falling within the general definition of confectionery even when that product is intended to meet the special nutritional needs of athletes."
- The Commissioners' published manuals confirm that, "Products intended to meet the specialist needs of athletes are not excluded from being considered as confectionery, but even though they may be sweetened, prepared and normally eaten with the fingers, they must also have the appearance and taste of confectionery." Volume V1-7 of those manuals, Chapter 1 – Food, section 4.7, indicates that traditional flapjacks are zero-rated as cakes. Flapjacks are described in the guidance in the following terms:
"Flapjacks: Traditional flapjacks have three main ingredients, oats, fat and sugar or syrup. Minor ingredients may be added to these, such as dried fruit, ground ginger and coating of chocolate, carob, yoghurt etc. which adds variety and do not significantly change the characteristics of the product. However, the inclusion of the other cereals would mean that the product is not a flapjack and as an item of sweetened prepared food eaten with the fingers, it would then be confectionery and standard rated. The texture of a true flapjack is cake-like, soft, moist and friable, not able to be snapped and not crisp. Traditional flapjacks are also usually rectangular shaped slabs, where as those flapjack-type products which are confectionery are usually thin and long (see also paragraph 4.11c)."
- Paragraph 4.11 (a) explains that:
"General: From 1st May 1988, Note 5 to Group 1 defines the term confectionery to include "any item of sweetened prepared food which is normally eaten with the fingers". This definition was incorporated to solve the liability problems raised by cereal bars, which were a new product at the time and did not easily fall into the working of the existing law, leading Tribunals to give contradictory decisions on very similar products. The most important point to bear in mind is that to be covered by this definition, products must have had sweetening matter added and must also be sweet to the taste. Food which is naturally sweet does not come within the definition, nor does food which has to be sweetened as part of the manufacturing process but is not sweet to the taste."
- Paragraph 4.11 (c) states that:
"Cereal bars: After 1st May 1988 and the amendment of Note 5, cereal bars have been standard rated as items of sweetened prepared food normally eaten with the fingers. There are products that appear to be similar to cereal bars. These include the traditional flapjacks, which are zero-rated as cakes. More recently there have been new products developed that have the appearance of cereal bars in terms of marketing, packaging and size, but have a soft, cake-like texture. These have also been zero-rated as cakes."
- Based on this guidance, the Commissioners concluded that Torq Bars could not be categorised as flapjacks and therefore as cakes. In particular, the inclusion of puffed rice as an additional cereal product removed Torq Bars from the category of a 'flapjack' within this relevant guidance.
The Appellant's contentions
- Mr Mills for the Appellant submitted that the correct approach was to give the words of the statute their ordinary meaning in their particular context. What is relevant is the view of the ordinary reasonable man in the street. In other words, what would be the view of the ordinary person as to the nature of the product. Does the product fall within the relevant category – in this case cake – or not? He drew our attention to the approach adopted by the Tribunal in United Biscuits (UK) Ltd v CCE (MAN/02/0534) VAT Decision 18596 where it had to address a similar question as to whether a particular product should be categorised as a 'biscuit':
"We bear in mind the comment made by Lord Woolf MR in Customs & Excise Commissioners v Ferrero UK Limited [1997] STC 881 at 884 that we should not address the issue in an over elaborate manner, but should concentrate on the single question, whether the product is or is not a biscuit; and we bear in mind also his approval, at page 885, of the comment made by the tribunal in that case that "we must give the word biscuit its ordinary meaning and must ask what view would be taken by the ordinary man in the street, who had been informed as we have been informed." Shortly put, the only issue before us is whether such a man in the street would regard this product as a biscuit. It is not necessary for us to determine what the product is if it is not a biscuit since, as Mr Gibbon accepted, if it is not a biscuit it simply falls into the residual class of confectionery which is standard-rated."
- The Tribunal then concluded that:
"While we accept Mr Poole's point that biscuits are not (yet) commonly eaten for breakfast we are not persuaded that his is the correct test. It seems to us to be dependant upon prejudice. The test in our view is not whether one eats biscuits for breakfast, but whether what one might eat for breakfast is a biscuit. Put that way, we do not think the man in the street would be unwilling to accept that this product was a biscuit. Although, as we have indicated, it is not necessary to find an alternative name for the product, the fact that – as we accept – none readily comes to mind does suggest that "biscuit" may not be an inappropriate term. We were impressed by the very wide range, in size, shape and superficial appearance, of the products which we saw and we are satisfied the term "biscuit" is not to be applied only to a narrow variety of product as might have been necessary 40 or 50 years ago. In our view, the informed man in the street would conclude that this product is entitled to the name "biscuit" and that the product should be so regarded."
- As to the nature of the product before us, Mr Mills said that the Torq Bar is a cake and should be zero-rated. He made two alternative submissions in support of this conclusion. First, he pointed out that flapjacks are categorised as cake and he said that Torq Bars are either flapjacks or are sufficiently similar to a flapjack for it to be categorised in the same way as flapjack. Alternatively, he said that soft, cake-like cereal bars are cake and Torq Bars are soft, cake-like cereal bars.
- Mr Mills referred to the tests that the Tribunal had applied in United Biscuits (UK) Ltd (No 2) (LON/91/160) VAT Decision 6344 to determine whether Jaffa Cakes were cakes or biscuits. In that case the Tribunal had identified 11 tests of cake. Applying those 11 tests to Torq Bars in the same way as the Tribunal had approached Jaffa Cakes in United Biscuits, Mr Mills commented as follows:
1. Name: This was regarded as a very minor consideration indeed.
2. Ingredients: In the present case Mr Mills said that the ingredients of the Torq Bar were sufficiently close to that of a flapjack for it to be considered as such.
3. Texture: The Tribunal in United Biscuits considered that, generally, it would expect a cake to be entirely or mainly soft and friable, not able to be snapped and not crisp. The Torq Bar was entirely or mainly soft and friable, cannot be snapped and is not crisp.
4. Size: Mr Mills said that the Torq Bar is the size of a small cake.
5. Packaging: Torq Bars are individually wrapped in foil like many cakes.
6. Marketing: Torq Bars are not sold in Supermarkets with cake but in sports shops and off the web site.
7. Manufacturing technique: Torq Bars are manufactured in a very similar way to flapjacks.
8. Consistency when stale: Like cake, Torq Bars are moist to start with and become harder and crisper when stale.
9. Presentation: Torq Bars are presented in a manner similar to many cake products.
10. Attractiveness to children: Torq Bars are designed to be attractive to adults, not children. In this way they resemble cake.
11. Core ingredients: Oats are a substantial part of the product, not in flavour but in bulk and texture when eaten. In this way the Torq Bar resembles a flapjack.
- Mr Mills noted that in United Biscuits the tribunal had found that Jaffa Cakes did not satisfy the test of cake under points 4, 5, 6, 9 and 10 above and decided that the tests in points 1 and 7 were unimportant. The Tribunal nevertheless decided that it was sufficient that Jaffa Cakes had the characteristics identified in points 2, 3, 8 and 11 for the Tribunal to conclude that Jaffa Cakes were cake. In this case, Torq Bars met the majority of these tests.
- Mr Mills conceded that the Torq Bar was an item of confectionery, in that it is a product which is sweetened and intended to be eaten with the fingers. This, however, was the characteristic of all cakes and biscuits. Mr Mills noted that the texture of Torq Bars is soft and cake-like. He said that the Commissioners had conceded in the correspondence that Torq Bars exhibit the characteristics of cake. They had stated that their inspection had indicated, "a soft and 'cake-like' product, which is moist, not crisp and can neither be easily crumbled or (sic) snapped." He submitted that the addition of other ingredients – such as puffed rice – were minor and for variation and did not alter the character of the product. If they were not cake as such, the consistency and texture of Torq Bars made them cake bars, which would also be zero rated under the relevant legislation.
- Mr Mills noted that the term "flapjacks" did not appear in the legislation. Nevertheless, the Commissioners considered that "flapjacks" had sufficient characteristics as to categorise them as cakes for these purposes. In the same way, Torq Bars had sufficient characteristics for them to be categorised as cakes in the same way as flapjacks. Torq Bars did not have to be flapjacks as the Commissioners had chosen to describe them in their literature.
- Nevertheless, Mr Mills criticised the way in which the Commissioners had sought to distinguish Torq Bars from traditional flapjacks. The Commissioners had asserted that 63% of the Torq Bar was made up of additional ingredients to what would traditionally be found in a flapjack. They said that this changed the character of the Bar. In arriving at this conclusion the Commissioners had relied on the absence of large quantities of fat in the Torq Bar's ingredients and had included both the fructose syrup and the maltodextrin (the "fat substitute") as additional ingredients.
- Mr Mills accepted that flapjacks usually had a high fat content and that this might be regarded as a traditional ingredient. He pointed out the evidence on the use of maltodextrin to replace fat. It had similar qualities to fat as a flavour enhancer like fat and giving a fat-feel to the tongue. He also noted that flapjacks are still flapjacks even when they are made with reduced quantities of fat for health reasons. He drew our attention to section 4.5 of Notice 701/14 which states that, "Low-calorie foods designed for slimmers are treated in the same way as their mainstream food equivalents."
- Mr Mills said the fructose syrup was a simple carbohydrate that can be considered as replacing the sugar in a traditional flapjack. Of the remaining products, 12.4% is fruit which differentiates one variety of product from another, 19.3% is liquidised raisins which form a syrup-like consistency and 7.4% is puffed rice. The fact that there is a second cereal product (puffed rice) did not in his submission affect the comparison with flapjacks.
- Mr Mills noted that the Commissioners' own guidance accepted that products that have the appearance of cereal bars in terms of marketing, packaging, and size but have a soft cake-like texture were zero-rated as cakes. Once they had accepted that Torq Bars had a soft cake-like texture it was unclear why the Commissioners considered that Torq Bars should not be zero rated as a soft cake-like cereal bar, even if they were not traditional flapjacks.
- Mr Mills accepted that the word "flapjack" did not appear on the advertising material or on the bar's wrapper, that the Appellant did not market its product as a flapjack or cake bar and that currently Torq Bars were not sold alongside a wide range of similar items in supermarkets. They were specialist cake bars currently supplied to a small market but Mr Mills said that if Torq Bars were sold in supermarkets they would naturally fit alongside flapjacks and breakfast bars rather than confectionery or biscuits.
- Mr Mills also said that in reaching our decision we should take into account the principles of Community VAT law. He said that we should be guided by the principle of fiscal neutrality. The tax should not be applied with the result that similar goods in competition with one another are treated differently (Albert Ruckdeschel & Co and Hansa-Largerhaus Stroh & Co v Hauptzollamt Itzehoe [1977] ECR 1753; Ideal Tourisme SA v Belgian State [2001] STC 1386). Although Member States may determine and define the transactions to which a reduced rate of VAT applies, in exercising that power Member States must respect the principle of fiscal neutrality. Mr Mills submitted that we should interpret the domestic legislation in a purposive way that would give effect to the principle of fiscal neutrality. That principle precludes in particular treating similar goods, which are in competition with each other, differently for the purposes of VAT (Christine Urbing-Adam v Administration de l'enregistrement et des domaine (Case C-267/99) [2001] ECR I-7467).
- He noted that at present Torq Bars appeal to a small market which is price sensitive even though the products that are sold to that market appear to be priced at a premium. He said that the profit margins were slim and at present three of the 4 main competitor products were presently zero-rated for VAT purposes. The effect of categorising Torq Bars as a standard-rated for VAT purposes in a market where the majority of competing products are zero-rated would be to drive Torq Bars out of the market, thereby reducing consumer choice. We should therefore interpret the UK domestic legislation to give effect to the principle of fiscal neutrality and categorise Torq Bars as cake.
- He also noted that zero-rating of food and cake can only be maintained for clearly defined social reasons and for the benefit of the final consumer (Article 28(2)(a) EC Council Directive 77/388/EEC (the Sixth VAT Directive)). He submitted that the clearly defined social reason for zero-rating food (as against confectionery) is to encourage the consumption of food of a high nutritional content as against the consumption of food for other reasons. Torq bars are purchased exclusively for their nutritional value. Mr Mills argued that the video evidence demonstrated that the ordinary man in the street considered the Torq Bar to be a nutritional product similar to a breakfast bar. It was not similar to traditional confectionery. For this reason also we should interpret the UK domestic legislation to give effect to the clearly defined social reason and categorise Torq Bars as cake.
- Finally, Mr Mills submitted that the legislative language should also be viewed in its context and having regard to its origins. He traced the origin of the excepted items in Group 1 of Schedule 8 to the Finance Act 1962 and then through the various emanations of the VAT legislation. He noted that Note 5 of Group 1 of Schedule 8 had been introduced following the decision in Customs & Excise Commissioners v Quaker Oats Ltd [1987] STC 683. In that case Quaker Oats had successfully argued that Harvest chewy bars were more akin to biscuits than to chocolates or sweets and therefore should be zero-rated. He submitted that the effect of the introduction of Note 5 was to change the question that we should ask ourselves from "is the relevant product more akin to cake than to sweets and chocolate" to "Does the relevant product have sufficient characteristics of cake to be classified as cake?"
- He further submitted that to the extent that it was possible to identify the intention of Parliament from reading the excepted items as a whole, one might consider that the excepted items are the sort of items which might, in 1962, be purchased with pocket money in a sweet shop and in that way could be distinguished from more nutritional foods which would be purchased from a grocery shop. He suggested that the development of the list of excepted items indicated that the intention of Parliament was to zero-rate food purchased for its nutritional qualities but to tax items of food not purchased for the purpose of nutrition. This would also accord with the clearly defined social reasons for maintaining a zero-rate of VAT on food as required by Article 28(2)(a).
The Respondents' contentions
- Ms Taylor for the Commissioners accepted that the zero-rating of food had a clearly defined social reason and benefited the final consumer. She said, however, that the rationale behind the legislation in issue was to allow zero-rating for most food and drink but to tax those items which might be considered non-essential. The wording of Note (5) had been changed to include within the definition of confectionery items of sweetened prepared food which were normally eaten with the fingers (see VAT (Confectionary) Order 1988 SI 1988/507) and this was primarily to include cereal bars within the excepted items. Accordingly she submitted that the Torq Bar was an item of confectionary and fell within excepted item 2 and was standard rated.
- She accepted that equality of treatment was a fundamental principle of Community law and that similar situations should not be treated differently without objective justification. National legislation should be interpreted where possible to give effect to the principles of equal treatment and fiscal neutrality so that similar goods that are in competition are not treated differently. She submitted, however, that this did not allow us to rewrite national legislation.
- She submitted that the Commissioners rulings and Tribunal determinations in respect of other products are of limited, if any, assistance to the Tribunal in determining the proper tax treatment of the Torq Bar. Each product had to be looked at on the basis of its individual merits. Following the principles of Community law, she said that the exceptions to zero-rating should not be narrowly construed but rather they must not be interpreted strictly.
- She said that the term "confectionery" in item 2 was clearly used in a wide sense. It included both "sugar confectionary", for example sweets and chocolates, and "baker's confectionery", for example cakes and biscuits. In this case it was plain that Torq Bars were an item of confectionery. They were intended to be eaten with the fingers and the addition of fructose syrup, which made up 14.9% of the Bar's overall content, was clearly "sweetening". The Appellant's own advertising material referred to the sweet taste of the bar, as did the reviews.
- She said that the next question was whether Torq bars could be categorised as a cake or a biscuit. The correct approach in considering whether a product is a cake, biscuit or other type of confectionary appeared in Ferrero UK Ltd (LON/94/1149) VAT Decision 13493, which had been approved by the Court of Appeal [1997] STC 881, and in United Biscuits (UK) Ltd (MAN/02/0534). These established that—
(1) words and phrases must be given their ordinary meaning;
(2) the question at issue is what view the ordinary man in the street would take of the product;
(3) it is irrelevant or inappropriate to consider the VAT treatment of other products;
(4) if a product has characteristics of more than one category, it may be placed in the category in which it has sufficient characteristics to qualify, i.e. as long as it has sufficient characteristics of the product to which it is to be classified it can be placed in the category to which it is akin.
- Taking the factors that the Tribunal had identified in United Biscuits (UK) Ltd (No 2) (LON/91/160), she said that Torq Bars did not have the appearance of a biscuit. They were a long thin rectangular shape, of a chewy, rather than a crisp, consistency and did not crumble. They did not snap when broken. Biscuits were generally made from wheat flour, fat and sugar combined to give a stiff dough which was cut and/or rolled for baking. Torq Bars by contrast were made of 22.3% oats, 14.8% fructose syrup and a variety of other ingredients but contained no fat. The dry ingredients were combined with a puree and pinned out into trays for baking. No "stiff dough" was formed from which the product was cut and/or rolled. She said that the Appellant did not market Torq Bars as biscuits and that the ordinary man in the street would not consider the product to be a biscuit.
- Similarly, she said Torq Bars were not cake. Cakes were usually made from a thin batter consisting of flour and eggs and were aerated in the process of cooking. Torq Bars do not have the appearance of a cake nor do they contain the typical ingredients of cake. The method used to produce Torq Bars was dissimilar to that used to produce cakes. The Appellant did not market Torq Bars as cake and she submitted that the ordinary man would not view Torq Bars as cake.
- Finally, Torq Bars were not flapjacks. She said that traditional flapjacks are cake-like in texture. They are soft, moist and friable and are not capable of being snapped. They are not crisp and are usually shaped in rectangular slabs. Traditional flapjacks have three main ingredients: oats, fat and sugar or syrup. Minor ingredients are sometimes included to add flavour and variety but these did not significantly affect the characteristics of the product. By contrast, the additional ingredients of Torq Bars represented 63% of the Bar. They were not simply minor additions for taste or flavour but change the character of the bar. Torq Bars did not have the appearance or texture of a traditional flapjack. They were long and thin in shape. They contained no fat, sugar or syrup but rather used fructose syrup as a sweetener. They included puffed rice as an additional cereal, which she submitted precluded their classification as flapjacks. She noted that Torq Bars are not described as a flapjack nor are they marketed as such. The ordinary man in the street would not view Torq Bars as flapjacks. Torq Bars were not sold with a wide range of similar items in, for example, a supermarket. They were marketed and sold as sports energy bars to improve athletes' performance when competing. Such bars were not offered any special relief under the legislation and should be standard rated.
- She submitted that it was irrelevant to consider how other products, such as the Kellogg's NutriGrain bar, the McVities HobNobs flapjacks with milk chocolate, the WeightWatchers flapjack, the Breakfast Tracker bar and the Sainsbury's chewy & crispy cereal bar, were categorised. In particular, we had no evidence as to how such products should be correctly categorised for the purposes of VAT and she said that it would be inappropriate for the Commissioners to express any view as to the liability of such products because this was a confidential matter as between the Commissioners and individual traders. The only relevance that these other products could have to the present proceedings was by way of limited comparison in terms of the factors that should be examined to determine liability of Torq Bars. Given the paucity of information available in relation to these other products and the fact that they were not truly similar to Torq Bars she said that we should give little weight to such evidence.
- She accepted that the High-5 Energy Bar and the SiS GO-Bar were similar to Torq Bars but did not accept that they were sufficiently similar to justify our giving significant weight to the comparison that had been drawn between these products and Torq Bars. In any event, the categorisation by this Tribunal of the SiS GO-Bar as zero-rated was on the basis that the product was not "sweetened" and therefore fell outside the excepted items to Group 1 of Schedule 8. This was not the case for Torq Bars.
The reasons for our decision
- We start by considering the Community law issues that the Appellant has raised. The zero-rating of supplies of food has clearly defined social reasons and also benefits the final consumer (see Procter & Gamble UK (LON/02/0896) VAT Decision 18381 at paragraph 41). We do not agree, however, with the way in which the Appellant seeks to express the social reasons. The majority of foods, if not all foods, including those that are clearly within the excepted items, provide some degree of nutrition and we do not think that the Appellant's attempt to draw a distinction for these purposes between what may be thought of as "good" and "bad" foods is soundly based. Many types of food that are commonly referred to as "junk" food qualify for zero-rating without ever getting into the issues raised by item 2 of the excepted items.
- If there is a common theme that lies behind some of the excepted items, it seems more likely to be that of "snack" foods as compared with foods that are more usually associated with one of the regular meals that may be catered for during the day, whether breakfast, lunch, tea, supper or dinner. To that extent Torq Bars are essentially snack food while, if it is a relevant point of differentiation, traditional flapjacks are as likely to be associated with the convention of afternoon tea as they are as snack food. We think that it is unnecessary, however, to express any final view on this point. It suffices to say that we reject the Appellant's arguments based on social purpose.
- The principle of equal treatment is, however, an established principle of Community law. This was accepted by the Tribunal in Procter & Gamble UK and we agree with the conclusion in paragraph 45 of its decision—
"From the authorities cited to us we derive the principle that equality of treatment is a principle of Community law. We cannot rewrite our national legislation in order to remove differences in treatment and, if such a rewriting were required, it would have to be in the direction of taxing all similar food products rather than zero-rating them. However, we can interpret the words of our national legislation if possible in such a way as to give effect to the Community principle of equal treatment, so that similar goods, which are in competition with each other, are not treated differently for the purposes of value added tax."
- As the Tribunal in Procter & Gamble UK had previously noted (paragraphs 41 and 42), zero-rating (and exceptions from the tax) are exceptions to the principle of the general tax on consumption and so have to be strictly construed. Furthermore, an exception to an exemption, because it subjects a transaction to the general rule that VAT is to be charged on all taxable transactions, should not be interpreted strictly. Thus, it does not appear to us that we are bound to construe the exception to zero-rating for confectionery narrowly. Conversely, we should construe the exception for cakes and biscuits (zero-rating those items notwithstanding the standard-rating of confectionery) narrowly; or at least should not aim to extend it beyond its natural limits. At the same time, while we accept the principle of equal treatment is one to which we ought to have regard, its application raises the question of which goods are similar and in competition to Torq Bars.
- From the evidence the main products in competition with Torq Bars are High 5 Energy Bars and SiS GO-Bars. The article from Cycling Plus magazine examines a number of other products, mainly different cereal bars but also Mars Bars and Jacob's fig rolls. Notably, the article does not include flapjacks. No doubt a cyclist could choose a cake, cake bar, Jaffa Cakes, flapjacks and a variety of biscuits, chocolate covered or otherwise, as refreshment and as a source of energy for a cycling event. Some of those products are undoubtedly zero-rated and some standard-rated. What they all have in common is that they are a source of energy. But that is true of many different foods – including ordinary bananas – and we do not think that it can be the relevant feature that determines whether we are dealing with similar goods in competition with each other.
- So far as its immediate competitor products are concerned, SiS GO-Bars are zero-rated because this Tribunal has decided that they are not confectionery. We do not know why High 5 Energy Bars are zero-rated (assuming indeed that they are) but given that the Appellant has conceded that Torq Bars are confectionery, the relevant question for Torq Bars and for similar products in competition with it is whether it is cake or a biscuit? We accordingly do not think that the principle of equal treatment advances the Appellant's case to any extent. It cannot operate to turn what is not a cake into a cake.
- We have also derived little assistance from the historical development of these provisions. We think it clear, as the Commissioners say, that the object of the change to the excepted items in 1988 was to deal with the Tribunal's decision in Customs and Excise Commissioners v Quaker Oats Ltd [1987] STC 683. Given the change in the statutory language, the nature of the question that we must answer has obviously changed, as Mr Mills contended. We do not think, however, that that takes the Appellant much further forward. Subsequent decisions of the Tribunal and the Courts have indicated quite clearly how we should approach the matter, as both parties recognised in their references to Ferrero UK Ltd v CCE LON/94/1149 and United Biscuits (UK) Ltd (AN/02/0534) VAT decision 18596.
- We are mindful of what the Court of Appeal said in Customs and Excise Commissioners v Ferrero UK Ltd [1997] STC 881, that we should not over elaborate matters and should seek to answer the question posed by the legislation: in this case whether Torq Bars can properly be described as cake. In answering that question we must consider the view of the ordinary person and what he or she would consider to be the nature of the product and whether or not it is one that falls within the relevant category, in this case as a cake. Where a product has the characteristics of two categories, it should be placed within the category in which it seems to fit most comfortably.
- We think it obvious that Torq Bars cannot be described as cake. We think that this is also true of flapjacks. On entering a tea room that offers a variety of products to consume with a cup of tea, one can envisage two questions coming to mind: the first is, shall I have a cake or a flapjack and the second is, if a cake, which type of cake shall it be? But as the Tribunal noted in United Biscuits (UK) Ltd (No.2) (LON/91/160), name (or description) is a minor matter.
- We have to consider, therefore, whether Torq Bars have sufficient characteristics of a cake to make them cake rather than some other product. If the question is whether the product is one of two products referred to in the legislation – a cake or a biscuit – it is possible to compare the two and decide into which category the product in question falls. On that basis, one might well be tempted to say that flapjacks, or certain flapjacks, have closer affinities to certain types of biscuit than to cake – oatmeal and raisin cookies, for example, which are frequently found to be soft and chewy depending upon the fat content and cooking method. Be that as it may, if the relevant question is whether Torq Bars have closer affinities to sports or cereal bars or to cake, the answer is obvious. But it does not answer the question whether or not sports bars, or particular types of sports bars, are cake.
- In the present case, the parties in their contentions have set out their respective views, which we have recorded in detail and need not repeat here. The Appellant seeks to get around the problem of categorisation principally by comparing Torq Bars not with cake but with flapjacks. In this respect we have not found the Commissioners attempts to distinguish flapjacks from Torq Bars entirely convincing. It does not seem to us that Torq Bars are so fundamentally different from flapjacks in terms of ingredients and production method as to make it impossible to conclude that Torq Bars are indeed a "healthy" or "low fat" form of flapjack. It may be the case that the relevant characteristics of Torq Bars are further removed from "traditional" flapjacks than they are from certain forms of cereal bar. On the other hand, there may be certain products that people would describe as flapjacks that could be compared to Torq Bars.
- These observations do not, however, answer the relevant question. The issue is not whether Torq Bars can be categorised as flapjacks but as cake. Flapjacks (or certain "traditional" flapjacks) may be sufficiently close in characteristics to cake to allow flapjacks to be categorised as cake. The fact that another product – in this case Torq Bars – is sufficiently similar in character to flapjacks to allow it to be categorised with flapjacks (if there were such a category) does not mean that it can be categorised as cake. The question always is whether Torq Bars have sufficient affinities with cake to be categorised as cake.
- Furthermore, with reference to United Biscuits (UK) Ltd (No.2) (LON/91/160), we have had regard to the characteristics of the product (Jaffa Cakes) that were taken into account there. We think, however, that we should avoid treating the particular way in which that the Tribunal in that case chose to describe each aspect of a Jaffa Cake as if it were a description of principles to be applied in each and every case. It was the Tribunal's choice of description based on the product involved and the question that the Tribunal has to answer by reference to that product. Thus in any particular case a Tribunal may choose to focus upon and mention particular aspects of the product – its ingredients, texture, size, packaging and marketing – but we do not think that the choice of description in one case should be treated as decisive in another. Cakes and biscuits come in a wide variety of ingredients, textures, sizes and packaging and may be marketed in a number of different ways. The final decision in any particular case is likely to be a matter of impression based on any number of different combinations of those aspects (and possibly others) rather than just seeing how many ticks there are on the list in comparison to the number of ticks on the particular list compiled by another Tribunal in another case.
- We have had regard to the manner in which the Commissioners have sought to describe the essential characteristics of traditional flapjacks in reaching the conclusion that they can be categorised as cake. We do not find their description particular convincing as a way of reaching this conclusion. The description of the texture of a flapjack as "cake-like" seems particularly self-serving. That is not to say whether flapjacks in general or traditional flapjacks in particular can properly be categorised as cakes or biscuits for these purposes. The fact, however, that Torq Bars may be "soft, moist and friable, not able to be snapped and not crisp" does not seem to us to suffice. We certainly would not describe its texture as "cake-like", notwithstanding the wide variety of forms that cakes can take.
- Accordingly, if (as we believe) the correct question is whether Torq Bars are cake, our answer on the basis of the evidence and our examination of the product is that they are not. Cakes certainly can differ widely and in approaching words descriptive of particular products – such as "cakes or biscuit" – we accept that we should be prepared to be flexible to allow for the development of new products within that particular category. Nevertheless, when all is said and done, there are limits even to that.
- It is true that one of the Torq Bars principal competitors, the SiS GO-bar, is zero-rated following an earlier decision of this Tribunal (SIS (Sport in Science) Ltd (MAN/00/69) VAT Decision 17116. The conclusion in that case was that—
"On the basis of the evidence presented to us, we find SiS's GO-BARs not to be sweetened prepared food, so that we hold them to qualify for zero-rating for VAT purposes: the main ingredients of each bar is sufficiently sweet to make it unnecessary for the bar to be sweetened."
- This seems to us to be an important point of differentiation. It may not be an easy point of differentiation to justify in any particular case but if sports bars such as Torq Bars and its competitors are to be zero-rated it seems to us that it must be because they are not confectionery rather than because they fall within the category of cakes or biscuits. As the Appellant has conceded that Torq Bars are items of sweetened prepared food, we must therefore dismiss its appeal and hold that they are standard-rated confectionery not being cake.
- In so holding we have reached the same conclusion on Torq Bars as the Tribunal reached on the product in issue in EJ Huczek (MAN/92/507) VAT Decision 8850. Mr Mills for the Appellant said that the main thrust of the taxpayer's argument in Huczek was that specialist sports foods should be zero rated per se. The taxpayer had not sought to say that its product was cake, nor had the taxpayer advanced any arguments based on the nutritional qualities of the product or sought to place the relevant UK legislation in its European legal context. For all these reasons he said that Huczek afforded little assistance. Be that as it may, our decision in this case is not based on Huczek but is based on the evidence about Torq Bars and whether that evidence leads us to believe that they can properly be categorised as cake. In our view they cannot be categorised as cake and we therefore dismiss this appeal.
MALCOLM GAMMIE QC
CHAIRMAN
RELEASED: 20 June 2005
LON/04/0205