[2016] UKFTT 302 (TC)

[image removed]

TC05071

 

Appeal number:  TC/2013/6480

 

VAT – food zero rating – whether supplies pre October 2012 heated for the purposes of enabling food to be consumed at a temperature above the ambient air temperature – whether supplies after September 2012 kept hot after being heated – preliminary issue decided in favour of HMRC

 

 

FIRST-TIER TRIBUNAL

TAX CHAMBER

 

	MUCHO MAS LIMITED T/A CHILANGO	Appellant
	- and -
	THE COMMISSIONERS FOR HER MAJESTY’S	Respondents
	REVENUE & CUSTOMS

 

 

TRIBUNAL:

JUDGE Barbara Mosedale

 

 

Sitting in public at the Royal Courts of Justice, the Strand, London on 1 & 2 March 2016

 

 

Mr T Brown, Counsel, instructed by RSM Tenon, for the Appellant

 

Mr E West, instructed by the General Counsel and Solicitor to HM Revenue and Customs, for the Respondents

 

 

 

 

 

 

 

© CROWN COPYRIGHT 2016

DECISION in PRINCIPLE

 

 

1.              The appellant sells prepared food from retail outlets located in London (five sites), Sheffield and Bluewater.  While its premises do have limited facilities for eating in, the majority of its sales are of take-away food, and it is only the takeaway food sales which concern this Tribunal.

2.              The appellant originally accounted for VAT on its takeaway sales which it later decided it should have zero rated.  So on 15 February 2013 it made a claim to HMRC for refund of input tax of £681,623.75 (allegedly) overpaid in the periods 12/08 to 03/12; it also ceased to account for VAT on these supplies.  HMRC assessed it to VAT on those supplies in the next VAT period (06/12) on 25 January 2013 in the sum of £59,328, and in period 09/12 on 15 January 2013 (in the sum of £28,416.72 also refusing a repayment of £20,419).  On 18 April 2013, HMRC further assessed the appellant for VAT for the period 06/09 to 09/12 in the sum of £257,511 on the basis it had treated sales as zero rated which HMRC considered should have been standard rated. 

3.              In brief, HMRC’s view was that all of the appellant’s takeaway sales should have been standard rated while the appellant’s view was that all of its takeaway sales should have been zero rated.

4.              The appellant appealed all these decisions and HMRC carried out a review.  HMRC issued two review decisions both on 16 August 2013, upholding their earlier decisions, and the appellant appealed to the Tribunal against the review decisions.

5.              I was not asked to make any decisions on quantum but merely to decide in principle whether the disputed supplies were properly zero rated or standard rated.

The facts

The evidence

6.              The appellant’s witness was Mr Luis Castro, who was head of quality assurance at the appellant company.  He had been with the company since January 2008, which was a few months after it commenced trading.  His role was to be in charge of the food and its quality.  I accept his evidence as largely reliable as it was consistent and he appeared to give genuine consideration to the questions put to him. The very limited extent to which I did not accept it is as set out, and for the reasons given in, §73 and the last sentence of §72.

7.              HMRC’s witness was Officer Sean Clayton.  Following receipt of the appellant’s voluntary disclosure mentioned at §2 above, he made various visits, together with Officer Norman Clark, to the appellant’s premises and undertook test purchases.  I accepted his evidence as reliable as it was also consistent and he appeared to give genuine consideration to the questions put to him. Indeed, while both witnesses approached the case from opposite ends, to the extent it was on the same topic, their evidence (but not their opinion on how it should be interpreted) was largely consistent with each others’.

Findings of Facts

8.              The appellant’s aim is to sell healthy, Mexican-style, fast food.  All its takeaway sales are in issue.  I find it sells five different kinds of product:

(1)          Burritos (accounting for 65% of sales)

(2)          nudos

(3)          totopos

(4)          tacos

(5)          salads

Burritos

9.              A burrito was prepared by a member of the appellant’s staff heating a round wheat tortilla on a hot plate for a few seconds in order to make it pliable.  The fillings chosen by the customer were then placed on the centre of the tortilla which was then folded or wrapped.  Once wrapped, the burrito was sealed in foil for the customer to take away.

Tacos

10.           A taco was very similar to a burrito, but made with smaller round wheat tortillas. They were sold in threes. The three tortillas were folded in half and the customer’s chosen fillings were placed on them and the tortillas were then rolled up.  This left some of the filling visible.  The three tacos would then be wrapped together in foil and given to the customer to take away.

Nudos

11.           A ‘nudos’ or ‘desnudo’, was prepared by a member of the appellant’s staff placing rice in a takeaway box, and placing the customer’s chosen fillings on top of the rice.  The customer then took away the closed box containing the nudos.

Totopos

12.           A totopo was prepared by a member of the appellant’s staff placing small corn chips in a takeaway box, and then placing the customer’s chosen fillings on top of the chips.  The customer then took away the closed box containing the totopos.

Salads

13.           A salad was prepared by a member of the appellant’s staff placing lettuce in a takeway box, and placing the customer’s chosen fillings on top of the leaves.  The customer then took away the box containing the salad.

The food line

14.           The customers chose the filings from the ‘food line’. The food line was a long stainless steel counter into which were inset stainless steel bowls containing the food.  Each bowl contained a different food item, eg chicken in one bowl, rice in another, and so on. The appellant’s staff prepared the food for the food line at the start of the day, and continued preparing it throughout the day as the need arose.  When prepared, the food was placed in the food line.

15.           Some of the bowls in the food line were kept hot, either by sitting in an outer bowl containing hot water (a bain marie) or by heating elements placed underneath. The customers could not see under the counter and so would not know what means were used to keep the contents of the inset bowls hot; however, I accept Mr Clayton’s evidence that a customer would in general know which items were hot on the basis steam could be seen to rise from the hot food items.

16.           The food line comprised the following choices:

(a)          grilled chicken (kept at 63ºC+);

(b)          steak (kept at 63ºC+);

(c)          pork (kept at 63ºC+);

(d)         prawns (kept at 63ºC+);

(e)          coriander rice (kept at 63ºC+);

(f)           black beans (kept at 63ºC+);

(g)          faitas (bell peppers sliced with onions) (kept at 63ºC+);

(h)          sour cream (unheated)

(i)            cheese (unheated)

(j)            guacamole (avocado based product) (unheated)

(k)          mild salsa (unheated)

(l)            medium salsa (unheated)

(m)        hot salsa (unheated)

17.           Each product was made to order; the customer decided whether he or she desired a burrito, tacos, totopos, nudos or salad, and then selected the fillings from the food line.  A typical selection would be a chicken burrito comprising chicken, rice, black beans, one of the salsas available, sour cream, cheese and with or without the optional extra of guacamole.  A typical vegetarian option would be a similar selection but with faitas instead of meat.

18.           Mr Castro accepted that although there were a vast number of possible combinations a customer could order, in fact most orders were one of a limited number of standard orders.  The appellant sold more chicken burritos than any other product.  It was possible, if unusual, for a customer to order one of the products containing nothing but unheated items (such as a burrito or salad containing only guacamole, cheese, cream and salsa).  I find most products sold contained at least one item kept at 63ºC+, and a standard order would include three (meat, rice and beans).

19.           I find that preparation of each customer’s order was very quick:  from the moment the customer indicated his choice to the member of staff to the moment the customer was handed the completed product would ordinarily be around a minute or less. 

The containers

20.           Mr Castro’s evidence was that the foil (used for burritos and tacos) was used as a means of preventing the product leaking onto their customers, and also because it adapted to the different sizes of the products sold.  I accept the appellant’s case that it could have used other packaging (such as paper combined with foil) which would have been a better insulator than foil on its own; nevertheless, as the appellant accepted, foil does have some ability to retain heat. 

21.           Mr Castro accepted that staff were shown a particular method of wrapping up tacos in foil.  It was his evidence that the objective of doing so was to prevent leaks but he accepted the same method would be used if the objective was to keep the tacos warm.

22.           Mr Clayton’s evidence was that during his 4 December 2012 visit to the appellant’s Upper Street premises, he overheard a staff member instructing other staff members to wrap Mr Clayton’s tacos in a particular way in order to keep them warm.  This is not hearsay evidence as HMRC did not rely on it as evidence of the truth of what Mr Clayton overheard but merely that as a matter of fact he did overhear what was said.  I accept his evidence. It is therefore clear, whatever Mr Castro’s views, at least some members of the appellant’s staff considered that keeping the product warm was an objective of the appellant’s.

23.           The boxes, used for nudos, totopos and salads, were ‘vegware’ made from a more readily compostable material than cardboard.  They comprised a single unit, incorporating a hinged lid.  They were described as ‘clam shells’.  They did not seal shut.

24.           I accept from the evidence on the containers that neither the foil nor the clam shell boxes were the best products for insulating hot food although both had some capacity to retain the heat of their contents.  I also accept that the appellant’s concerns with its choice of packaging included the ability to supply a product that did not leak and was not messy and the cost of the packaging. At the same time, taking into account what Mr Clayton overheard, I was not satisfied that the appellant had no interest in keeping warm the products handed over to its customers.

The temperatures

25.           Both HMRC and the appellant tested the temperature of the appellant’s products.

26.           Originally the appellant carried out tests in October 2012.  It tested 8 burritos, taking two or three temperatures from each burrito, at different points, and then averaged the temperature for each burrito.  Its results (8 average temperatures) ranged from 16.9ºC to 25.5ºC all of which were below the ambient air temperature (as to which see XXX below).

27.           After a meeting with HMRC, it carried out further tests on 3 burritos in December 2012, using a different method but again taking a temperature at three different points from the same burrito and obtaining an average.  The 11 December burrito varied between 13.4ºC and 37.2ºC producing an average of 23.8ºC; the first 12 December burrito varied between 13.5ºC and 48.2ºC producing an average of 27.7ºC; the second 12 December burrito varied between 12.3ºC and 37.4ºC producing an average of 24ºC.

28.           Each HMRC officer also conducted a test purchase on their Fleet Street visit on 16 November 2012; they inserted the thermometer, within a minute or so of purchase,  a number of times into each product but only recorded the highest temperature obtained.  For a grilled chicken burrito the temperature was 41ºC and for a grilled chicken tacos the temperature was 42ºC.  They then each carried out a test purchase at the appellant’s premises in Chancery Lane under similar conditions and obtained a maximum temperature of 33ºC for a grilled chicken salad and 58ºC for a grilled chicken totopos.  On 4 December 2012 they conducted test purchases under similar conditions at the appellant’s Upper Street premises and obtained maximum temperatures of 48ºC for a chicken burrito and 30ºC for a steak taco.

29.           I consider that the variation in temperatures between each product tested (30ºC to 58ºC) is to be expected and does not affect the veracity of the results.  The products are prepared from a range of ingredients some of which have been kept hot and some of which have not been kept hot; some customers would order products containing less hot ingredients than others. And the temperature of some hot ingredients might exceed 63ºC when added to the product (eg if just cooked/re-heated).  So it is to be expected that the products will very in temperature one from another and that each burrito will have had variation in its internal temperatures.

30.           Mr Brown suggested that the officer’s records of the temperatures was not reliable as:

(1)          Mr Clayton admitted he did not write them down until he returned to the office and wrote up his notes;

(2)          Mr Clayton had later made a mistake in a letter when transcribing the temperatures from his notebook; and

(3)          (the appellant said) HMRC’s thermometer was not calibrated properly and not calibrated at all immediately before each use. 

(4)          Mr Clayton had avoided cold items such as guacamole and salad in his choices.

31.           While I agree that Mr Clayton’s failure to immediately write down the temperatures might in some circumstances give cause for doubt about their accuracy, in this case not only was Mr Clayton sure he had written them down correctly but they were roughly consistent with the appellant’s maximum temperature readings as set out in §27.  For the same reasons, while I accept that Mr Clayton was careless in a letter he wrote, I see no reason to suppose he wrote down the temperatures inaccurately when they are in line with the appellant’s own measurements.  So I accept the temperatures recorded by Mr Clayton as accurately recorded. 

32.           So far as the accuracy of the thermometer was concerned, I find HMRC calibrated their thermometer by inserting it into cold water at the start of the day: if it responded by decreasing its temperature reading, HMRC would rely on it to take readings for the rest of the day.  The appellant’s view was that the thermometer should be calibrated between every use and should have been checked against boiling water.  Again, while I accept that the appellant’s suggested method of calibration would test not only whether the thermometer worked, but worked accurately, I have no reason to suppose that HMRC’s thermometer did not work accurately because the range of temperatures obtained was consistent with the appellant’s.  I accept them as accurate.

33.           Indeed, as Mr West said, it was the appellant’s evidence that the hot ingredients were kept at at least 63ºC and that, when a customer chose a product, the ingredients were assembled quickly.  So it would be very surprising if the hot part of the product was not above, often considerably above, ambient air temperature when it was handed to the customer. I also think it is irrelevant that Mr Clayton did not choose guacamole as an ingredient and a salad as a product as doing so could only have marginally reduced the temperature of the hot part of the product from what it was without these items. So I reject the appellant’s challenge to HMRC’s evidence on temperatures.

34.           There was little dispute over what was the ambient air temperature.  Both parties accepted it was the ambient air temperature of the appellant’s premises; HMRC accepted that the appellant set the heating to 28ºC although on the day of the first meeting between the parties on 11 December 2012 the actual room temperature was 26ºC.  It makes no difference to the outcome of this appeal whether the ambient air temperature was 28ºC or 26ºC and I will proceed on the basis it varied between 28 and 26ºC.

35.           There was no evidence directed at the temperature of the unheated items in the food line.  There was no suggestion that they were refrigerated while on display:  indeed it seems that they were merely placed in an unheated bowl in the food line. However, the accuracy of the appellant’s evidence on temperatures of its burritos was not contested: HMRC merely took the position, with which I agree, that average temperature was not relevant to Note 3(ii) (see §57-59 below).  But the appellant’s evidence on temperature (26-27) strongly suggests that the unheated items, or some of them, were below ambient air temperature at the time of supply.  This is because (a) the temperature of the cooler sections of the burritos ranged from 12.3 ºC to 13.5 ºC and (b) the average temperatures of the overall product ranged from 16.9 ºC to 27.7 ºC.  So it seems more likely than not that the unheated items, or some of them, were significantly below ambient air temperature at around 13ºC.  (The likely explanation is that they were refrigerated before being put on display but I had no evidence on this).

The origins of the burrito

36.           Mr Castro’s witness statement went into some details on what he believed from his personal research to be the origin of the burrito.  He considered it started out as a form of ‘packed lunch’ provided to field workers in Mexico.  Mr West put it to him that even if that was the origins of the burrito, it had no relevance to the food produced by the appellant.  Mr Castro disagreed.  He considered the appellant produced a form of packed lunch for office workers, who would have the same interest as agricultural workers in a tasty packed lunch containing a mixture of protein, vegetables and carbohydrate which would keep them going for the afternoon.

37.           The root of Mr West’s disagreement with Mr Castro’s evidence was Mr Castro’s evidence that the burrito was in historical times prepared at the start of the day but not eaten by the agricultural workers until lunchtime, by which time the burritos would be cold and congealed.  Mr West’s point was that Chilango did not sell cold, congealed burritos prepared some hours earlier:  the appellant’s burritos were prepared fresh to each customer’s order.

38.           There was no evidence that the appellant used Mr Castro’s understanding of the origins of the burrito in its advertising so there was also no suggestion from this that the appellant’s customers expected cold food. I find that the evidence about the historical burrito was not relevant.  I agree with Mr West that the evidence shows that the appellant prepared its products freshly to order and had the expectation that they would be eaten shortly afterwards as the appellant expected its customers to take their purchased food to consume as soon as they were back in their offices.

Advertising

39.           Mr Castro accepted that the appellant advertised its products as ‘fresh’ and referred to its meat products (depending on cooking method) as ‘braised’ or ‘grilled’.

Customer expectations

40.           I had no evidence from customers so their expectations are difficult to judge. Mr Castro gave evidence that a small percentage of customers complained the food was too cold and wanted to know how to warm it up.  That is evidence that a small minority of customers at least expected or wanted hot food. Mr Castro’s evidence is that the appellant’s staff would try to modify the expectation by explaining how the burrito was produced:  the appellant would not heat up the product.

The food hygiene regulations

41.           Strictly the food hygiene regulations are law but it makes sense to consider them in the ‘facts’ section as the question was not what the regulations were, but whether the appellant heated its food in order to comply with them or for a different motive.

42.           The law:  I was referred to the Food Hygiene (England) Regulations 2006 SI no 14.  Schedule 4 of these regulations dealt with temperature control requirements.  Paragraph 2 of Schedule 4 required food which could support the growth of ‘pathogenic micro-organisms’ or toxins to be kept by food retail outlets at 8ºC or less and if it was not an offence was committed.

43.           There were exceptions for cooked/reheated food on display for sale which was kept at or above 63ºC (paragraph 3(a)); there was also an exception for any food which could be kept at ambient air temperature without risk to health while on display for sale (paragraph 3(b)).

44.           It was an offence for food which needed to be kept above 63ºC (ie food within the paragraph 3(a) defence) to be kept at below 63ºC.  There were only two defences to this offence:  one was that the food was kept on display for sale for less than 2 hours (paragraph 7(2)) and the other where scientific assessment had been carried out that what was done involved no risk to health.  Neither party suggested that this last exception could or should have been utilised by the appellant and I don’t refer to it again. 

45.           Application of food hygiene rules in the appellant’s business:  Mr Castro gave evidence about chicken but I understood what he said applied to all the meat sold.  Chicken was purchased raw and delivered refrigerated at the start of each day.  Staff cooked the meat; it was cooked until it reached and maintained for 3 seconds the internal temperature of 75ºC as the appellant believed was required by food safety law.  Mr Castro’s evidence was that, as one would expect and I accept, meat did support the growth of pathogenic micro-organisms; in these circumstances the law, set out above, then gave the appellant a choice of how it treated cooked meat if it was to avoid committing an offence:

(a)          Do nothing and allow it cool but if so it must be sold within 2 hours;

(b)          Cool the product to 8ºC or less within a specified period and store it refrigerated;

(c)          Keep the food above 63ºC.

46.           Option (b) was obviously inappropriate to the appellant’s fast food business which used fresh ingredients each day and disposed of unused cooked food at the end of each day.  The appellant’s choice was option (c) to keep the meat hot at 63ºC or above.  Mr Castro accepted that at least over the busy period of lunch (12 noon to 2pm) the appellant was likely to sell meat placed on sale within two hours and theoretically option (a) was also open to it during that period.  However, I accept his evidence that sales were slower and less predictable outside the lunch period and the appellant was not certain to sell meat placed on sale within two hours outside the lunch period.  He considered it too complicated for staff, for many of whom English was not a first language, and many of whom only stayed in the job for 2 or 3 months, to operate option (a) over the lunch period and option (c) at all other times.  He considered it much safer to have simple rules for the staff to follow as that decreased the risk of staff getting it wrong and the appellant committing an offence and risking giving food poisoning to their customers.  The appellant’s instructions to staff, therefore, were that once cooked, meat had to be kept at 63ºC until the moment of sale.

47.           HMRC challenged this evidence.  Mr West pointed out that it was cheaper not to keep food hot when it did not need to be kept hot.  However, I accept the evidence that it was the appellant’s choice, for reasons of operational simplicity, to comply with food safety laws by maintaining its meat products at above 63ºC.

48.           I had a lack of detailed evidence about the other food items kept heated in the food line.  So far as beans were concerned, I had an explanation of why they were heated.  This was because beans were delivered pre-cooked and refrigerated (ie under 8ºC).  Beans at less than 8ºC are too viscous to divide into portions accurately and quickly and must therefore be served at least at room temperature.  The appellant therefore re-heated the beans.  I was given no explanation of why the rice and faitas were  heated.

49.           And I had no evidence about why the beans, rice and faitas were then kept hot in the food line.  In particular, I had no evidence from Mr Castro on whether these non-meat items were like chicken and supported the growth of pathogenic micro-organisms which, under para 2 of Sch 4 of the food hygiene regulations, would mean the appellant could only allow them to cool if sold within 2 hours.  I had no evidence whether, instead, the beans, rice and/or faitas could be kept at ambient air temperature without risk to health while on display for sale (Sch 4 para 3(b)).

The law

50.           The law applicable to the appellant’s sales changed with effect from 1 October 2012.  While, as can be seen from §2, the appellant’s initial position was that all five items in issue were zero rated both before and after the change in law, by the hearing the appellant had accepted that all its items bar the salads were standard rated after 1 October 2012.  To that extent it accepts its appeal must be dismissed in principle.

51.           I was asked to consider the proper VAT treatment of all five items up to 1 October 2012 and the proper VAT treatment of the salads on and after 1 October 2012.

52.           The relevant law was the provisions on the zero rating of food which was and still is contained in Group 1 of Schedule 8 to the Value Added Tax Act 1994. Group 1 provided so far as relevant that the supply of food is zero rated

“…except –

(a) as supply in the course of catering…”

And then Note (3) provided up to 1 October 2012:

“(3) a supply of anything in the course of catering includes –

(a) any supply of it for consumption on the premises on which it is supplied; and

(b)any supply of hot food for the consumption off those premises;

And for the purposes of paragraph (b) above ‘hot food’ means food which, or any part of which –

(i) has been heated for the purposes of enabling it to be consumed at a temperature above the ambient air temperature; and

(ii) is above that temperature at the time it is provided to the customer.”

53.           It was accepted that the takeaway food supplied by the appellant before 1 October 2012 would only be ‘hot food’ and therefore standard rated if conditions if both (i) and (ii) were met.

54.           After September 2012, the definition of ‘hot food’ changed and became:

(3B) ….food which (or any part of which) is hot at the time it is provided to the customer and –

(a) has been heated for the purposes of enabling it to be consumed hot;

(b) has been heated to order;

(c) has been kept hot after being heated;

(d) is provided to a customer in packaging that retains heat (whether or nto the packaging was primarily designed for that purpose) or in any other packaging that is specifically designed for hot food; or

(e) is advertised or marketed in a way that indicates that it is supplied hot.

(3C)  For the purposes of Note (3B) –

(a) something is ‘hot’ if it is at a temperature above the ambient air temperature; and

(b) something is ‘kept hot’ after being heated if the supplier stores it in an environment which provides, applies or retains heat, or takes other steps to ensure it remains hot or to slow down the natural cooling process.

(3D) In notes (3B) and (3C), references to food being heated include references to it being cooked or reheated.

55.           Again, there was a test in two parts, and the food had to fail both to be standard rated.  As before, although not numbered (ii), the food, or a part of it, had to be above ambient air temperature at the time it was provided to the customer to fail the first test.  The second test (equivalent to old (i)) was rather different:  while the criteria of old Note 3(i) ‘has been heated for the purposes of enabling it to be consumed hot’ was retained, new alternative criteria were added ((3B)(b) to (e)) above, making it easier for food to fail the test and be standard rated.

56.           I will first consider that part of the pre and post 1 October 2012 test which is common to both:  were the products, or any part of them, above ambient air temperature at the time they were provided to the customer?

Above the ambient air temperature at the time provided to customer?

57.           The appellant’s various tests described above were aimed at measuring the average temperature of burritos; the appellant did not agree that HMRC should take as the temperature the hottest part of the food.  The nature of the appellant’s five products was that each was a mixture of hot and cold ingredients.  The appellant thought an average temperature was more appropriate and fairer way to determine whether product as a whole was above the ambient air temperature at time provided to the customer.

58.           However, I agree with HMRC, that as a matter of law the test was whether any part of the product was above ambient air temperature at the time it was provided to the customer.  Note (3) specifically stated

‘… ‘hot food’ means food which, or any part of which – …’

And note (3B) in the new legislation also referred to ‘any part of which’.

59.           In conclusion, I am not asked to consider the average temperature of the product at the time it was provided to the customer; I am asked to consider if a part of the product was above ambient air temperature at the time it was provided to the customer.  While there may be an argument about de minimis, so that a cold item only of very small part of which was hot might nevertheless not be hot food, there was no suggestion that the ‘hot’ part of the products was de minimis here.  Mr Brown’s skeleton refers to less than 50% of each product being hot but I did not even have the evidence to reach that conclusion.  On the contrary, the evidence was the standard product was a chicken burrito comprising heated chicken, rice and beans, and unheated cheese, cream, salsa with optional guacolmole, with no evidence on respective quantities of each.  And even if the ‘less then 50%’ figure was correct, I am satisfied that the part of the product which gave the maximum temperatures referred to below at §61 was not a de minimis part of the product.  In conclusion, I will consider the maximum and not average temperatures obtained when the various products were tested.

60.           Therefore, I have to disregard the results of the appellant’s October 2012 tests as only the average temperature was recorded.  I also disregard the average temperature generated in the appellant’s later tests and consider only the maximum temperature obtained. 

61.           HMRC accepted the relevant ambient air temperature in this appeal was 26ºC-28ºC. In the tests for which a maximum temperature was obtained, a part of each of the 9 products tested (see §§27-28), including the 3 tested by the appellant,  were above this temperature.  The maximum temperatures of the 9 products ranged from 30ºC to 58ºC (the average maximum temperature was about 42ºC).

62.           While not every product could be tested, not least because each product was a combination of ingredients chosen by the customer and there were a very large number of possible combinations, I am satisfied that the sample was representative.  It included burritos, tacos, totopos and a salad.  No one suggested the results would be much different for a desnudo and indeed as the ingredients were the same (bar the base) logically I would not expect any significant difference.

63.           In conclusion, therefore, I find that a significant part of each of the appellant’s products at issue in this appeal (subject to the minor caveat in §§98-99) would have been above the ambient air temperature at the point that the products were provided to the customer.  Therefore, the products failed the temperature part of the test for zero rating both before and after 1 October 2012.

64.           I move on to consider the remaining part of the test in each of the two relevant periods.

The law before 1 October 2012

65.           There has been a great deal of litigation centering on the meaning of Note (3)(i).  The Court of Appeal decision in John Pimplett & Sons Ltd [1988] STC 358 decided that it was the supplier’s intention which mattered (eg page 361a-b):  in that case the Court decided that because the supplier heated the pies in order to make them fit for sale, and did not heat them for the purpose of enabling them to be consumed hot, the sale was zero rated.

66.           Thereafter, there have been many cases where there has been a dispute over the supplier’s intentions.  However, in Sub One Ltd (t/a Subway) ‘Sub One’ [2014] EWCA Civ 773 the Court of Appeal ruled that the test in Note (3)(i) was an objective test and Pimblett had been wrong to apply a subjective test, albeit one which could be assessed from objective factors (such as what the supplier actually did).  The Court of Appeal in Sub One said that the test in Note (3)(i)

‘[49]….searches for the assumed common intention of the supplier and the consumer as to whether it is a term of the bargain that the product be supplied in order to be eaten hot.  By this entirely objective enquiry, the court derives the terms of the bargain from what each party to the contract says and does (including presentation of the supply in the shop and in any advertising).’

67.           HMRC’s position was that all of the cases decided before Sub One were, by applying Pimblett, decided on an incorrect understanding of the legal test and should not be relied on or referred to.  Moreover, they were decided at Tribunal level in any event and not binding. The appellant accepted that the test was objective but disagreed that all previous case law was necessarily irrelevant.  I was referred to N M Holmes t/a The Chicken Shop MAN/98/355 where the Tribunal held that the supplier’s main purpose in heating the chickens and keeping them hot was to comply with health and safety laws and therefore the supply of them was zero rated.  My view is that this case is not really relevant because it is clear that the supplier’s actual prime purpose is no longer the test; while the supplier’s actual intention may well be relevant, the Tribunal must now find the assumed common intention of the supplier and consumer, which may not be the same as the supplier’s actual intention.  The Tribunal must find out if it was a term of the bargain for the food, or a part of it, to be supplied hot.

68.           I accept that, as Mr West said, the burden of proof is on the appellant.  The appellant must satisfy me it was not the common intention of the parties that the product was heated for the purposes of enabling it to be consumed at a temperature above the ambient air temperature.  In searching for the common intention, I have to consider both supplier’s and consumer’s actual intentions in so far as it is possible to discern them, but that is only in order to determine the common intention, which may not be the actual intention of either supplier or consumer.

The appellant’s actual intention?

69.           I am satisfied, relying on Mr Castro’s evidence summarised at §§45-47, that a primary purpose of the appellant in heating and then keeping warm the meat ingredients was to comply with food safety laws and indeed to ensure that the meat was safe to eat.

70.           Mr West suggested that the appellant had to comply with food safety laws in order to sell food fit for human consumption; had it not done so the food would not have been within Group 1 at all, so, said Mr West, it could not then also rely on such compliance to prove it was not a supply in the course of catering.  I do not follow the logic of this and I do not agree with it: there is nothing illogical in a taxpayer’s purpose both qualifying the food within Group 1 as fit for human consumption, and also meaning it was not a supply in the course of catering.  But this appeal does not turn on the supplier’s sole or main purpose but the common intention of buyer and seller and I move on to consider that below.

71.           Mr West also said that the appellant had to demonstrate that, but for the food hygiene regulations and the need to sell safe food, the appellant would have sold all the ingredients cold.  What I understood him to mean was that, if the appellant could not demonstrate this, I should infer there was another primary reason, not just food safety, for keeping the food hot. I agree that a dual purpose would be relevant but it is no longer conclusive as the question is the common intention, objectively determined, of both buyer and seller.

72.           Did the appellant have a dual purpose?  While I accept that the beans had to be heated, I did not have evidence why the rice and faitas were heated (§48):  while it is common knowledge rice must be cooked to be edible, I was not told whether the rice was purchased uncooked or precooked (like the beans); and  I also consider it common knowledge that red peppers and even onions don’t have to be cooked to be edible.  So I do not know why the rice and faitas were heated.  And I was given no evidence on why the beans, rice and faitas were kept hot once heated (§49).  As I do not know why these ingredients were kept hot, I cannot be satisfied that the main reason for doing so was food safety; I cannot even be satisfied that food safety was the only reason, rather than merely one of a number of prime reasons, why the meat was kept hot.  The appellant might have kept meat hot, even without food safety concerns, for the same reason (whatever it was) it kept the rice, beans and faitas hot.

73.           While I accept that the appellant did not use the most insulating form of packaging and that  heat retention was not the only motivating factor in choice of packaging, taking into account Mr Clayton’s evidence at §24, I was not satisfied that heat retention was not a consideration for the appellant in its choice of packaging.

The common intention?

74.           But that is far from the end of the matter.  It is not the supplier’s intention which decides the matter but whether it was the objectively determined common intention of supplier and consumer that the food was to be consumed above ambient air temperature.  When considering a common intention it seems to me that I must look at what was known to both parties, how they acted and what if anything was said eg in advertising.  An actual intention which was not communicated or known to the other party may not form part of the objectively determined common intention.

Advertising

75.           The only evidence on advertising was as set out above in §39.  HMRC’s case is that these words implied the food was hot and customers would understand that that was what those words implied.  I am unable to agree. Braising and grilling are cooking processes and I do not accept that a customer ought to or would understand the words to mean anything other than that the meat had been cooked by those processes.  The words did not imply that the meat would still be hot when eaten.  Similarly the word ‘fresh’ would import that the food was recently prepared from raw ingredients, which would imply recently cooked, but again would not necessarily imply the cooked food was still hot.

76.           There was in my view nothing in the use of these words that would contribute to any common understanding that the food was intended to be consumed hot.

Presentation of food

77.           In the absence of advertising, the common intention must be gleaned largely from how the food was presented to the customer in the food line and how it was packaged when handed to the customer.

78.           The appellant knew certain of its products were held at at least 63ºC.  Other ingredients were not held warm.  I find it would also have been apparent to the appellant’s customers that some ingredients were kept hot and some were not (§15).  It is clear that the appellant expected its customers to choose from both cold and hot items when making a selection of filings for any of its five product types (burritos, tacos, totopos, nudos and salads) and I find this would have been the customers’ expectations too and indeed the evidence is that this is what they did.  (It was very rare for customers to select just cold items §18).

79.           Ordinary and well known laws of physics mean that the hot items will rapidly cool when placed next to colder items and both the appellant and its customers must be taken to have known this.  Indeed, the evidence is that this is what happened:  the range of maximum temperatures showed that some items, served at at least 63ºC, had, in the short space of time it took for them to be served and then taken outside to be tested, sometimes cooled to as little as 30ºC, although the temperature reduction was not always as drastic (only to 58°C in one case).

80.           It was the appellant’s case that the entirety of the product would have cooled to ambient air temperature by the time it was eaten.  While the appellant expected its customers to eat the product shortly after purchase, it considered its customers would most likely take them back to their office and that it was its case a product which had cooled from say 63ºC to between 30ºC-58ºC in the first few minutes would, by the time it reached the customers’ place of work, be entirely at ambient air temperature.

81.           I accept that when the cold ingredients were combined with the hot ingredients there would have been initial rapid cooling of the hot ingredients but at the same time the temperature of the unheated ingredients must have increased and it must be obvious to both parties that this would happen. So even though I accept at least some of the cold items were below ambient air temperature at time of supply (§35), that does not tell me how long it would take for the hot ingredients to cool to no more than ambient air temperature (the test being whether ‘any part’ of the product ‘has been heated for the purposes of enabling it to be consumed at a temperature above the ambient air temperature’) and whether it was likely they would have done so by the time of consumption.  While the initial cooling was rapid, that does not mean the cooling would have continued at the same rate:  it is I would think common knowledge that the rate of cooling would decrease as the difference in temperature decreased: in other words, supplier and customers would or at least should expect heat exchange between the hot and cold items to be rapid at first but tail off as the items got closer in temperature.

82.           In conclusion, I consider that the appellant has not proved that every part of the products in issue were at ambient air temperature at point of consumption, nor that the appellant and customer objectively expected that they would be.  It seems more likely to me that the customer, choosing a number of hot and cold items and eating them as intended relatively soon after purchase, might well expect the hot items to still be above ambient air temperature at the time of consumption and the appellant ought to have expected this too.  Whether that objectively determined expectation was a part of the bargain between the parties I consider below.

Packaging

83.           The burritos and tacos were wrapped in foil and the other three products placed in a box.  It was clear that the products had to be wrapped in something in order to be taken away and eaten; ensuring that the packaging did not leak or allow the product to fall apart may well have been the appellant’s and customers’ prime concern with the packaging. I am inclined to think that the choice of foil, at least, might indicate to the customer that the appellant was seeking in its choice of packaging to aid the retention of heat.

84.           Mr West pointed out that Mr Castro’s evidence was in effect that a small minority of customers had an expectation that the products would be warmer than they actually were: as this evidence appeared to relate to a very small minority I do not think is indicative of what objectively the majority of the customer’s expectations were.  Indeed, as the majority did not complain about the temperature of the food, the evidence indicates that the vast majority were not dissatisfied with it.  But while this evidence therefore indicates that the expectations of the majority on temperature were met, it does not tell me what temperature the majority expected the products to be.  So this evidence does not help either HMRC’s or the appellant’s case 

Food safety laws

85.           What relevance are the food safety laws when one is considering the presumed common intent?  Customers should, I think, be presumed to have the common knowledge that there are food safety laws with which fast food retailers ought to comply, but I do not think that they can be presumed to know exactly what the rules are.  And even if they did, it would not be obvious to them that the meat was kept hot for reasons of hygiene because, as I have said, the appellant had two options in law (§46) and chose for reasons of commercial convenience to opt for keeping the meat at 63 ºC.  Moreover, I do not think the customers would assume that the other hot items were kept hot for reasons of food hygiene because I do not think it is common knowledge that they must be kept hot for reasons of food hygiene and indeed I do not know why they were kept hot (§49).  In conclusion, while I accept that it was a term of the bargain that the appellant would supply food that was both fit for human consumption and complied with food safety laws, I do not think, objectively determined, that the customers would consider that the food was kept hot in order to meet either of these conditions.

86.           I have also said that the appellant’s intention to keep meat hot in order to comply with food safety laws was not communicated to its customers.  So taking into account this purpose was neither communicated to the customer and nor do I think customers would have assumed keeping the hot items warm was for reasons of food safety, the appellant has not made out its case that it was the objectively determined common intention of the parties that those items kept hot were kept hot for food safety reasons.

Conclusion

87.           Mr Brown says the Tribunal had to ask itself whether the customer expected, having loaded his or her purchase with a mixture of hot and cold ingredients and having taken it back to his or her office to be consumed, that any part of the product would still be hot? I agree that the answer to this question is relevant to the issue the Tribunal must decide but I don’t reach the answer to the question that Mr Brown wants me to reach.

88.           It was not proved, and it was not obvious to me, that in these circumstances the heated elements of the product would no longer be above ambient air temperature at point of consumption.  They were likely no longer to be ‘hot’ in the sense it is commonly understood but it was not shown that they would not be ‘hot’ in the legislative sense of above ambient air temperature.  Moreover, I do not think this would be obvious to either the appellant or its customers.

89.           Of course, even thought  I have found (§82) that the customers must have expected parts of the product to still be warm (ie above ambient air temperature) at time of consumption that does not show there was a common intention or expectation that the heating was for the purpose of enabling them to be consumed at that temperature.  But as I have said, I have rejected the appellant’s case that the common intention or expectation was that the heating was for the purpose of food safety requirements.  So in lieu of that, what else would the common intention as to the purpose of the heating to be?

90.           It seems likely to me that the customer would assume it was for the purpose of eating hot (in the sense of above ambient air temperature) food.  In conclusion, I think that the appellant has failed to demonstrate that the products (containing all or some of meat, rice, beans and faitas) were not heated for the common purpose of enabling them, or a part of them, to be consumed at a temperature above the ambient air temperature.  Its appeal with regards the period before October 2012 must be dismissed.

The law after September 2012

91.           As I have said, the definition of ‘hot food’ changed and became:

(3B) ….food which (or any part of which) is hot at the time it is provided to the customer and –

(a) has been heated for the purposes of enabling it to be consumed hot;

(b) has been heated to order;

(c) has been kept hot after being heated;

(d) is provided to a customer in packaging that retains heat (whether or nto the packaging was primarily designed for that purpose) or in any other packaging that is specifically designed for hot food; or

(e) is advertised or marketed in a way that indicates that it is supplied hot.

(3C)  For the purposes of Note (3B) –

(a) something is ‘hot’ if it is at a temperature above the ambient air temperature; and

(b) something is ‘kept hot’ after being heated if the supplier stores it in an environment which provides, applies or retains heat, or takes other steps to ensure it remains hot or to slow down the natural cooling process.

(3D) In notes (3B) and (3C), references to food being heated include references to it being cooked or reheated.

Post September 2012 – salads

92.           As I have said, the appellant accepted that its supplies, except salads, post September 2012 were standard rated.  I do not really understand its position here.  It is clear, and was accepted, that conditions (a) to (e) of (3B) of Group 1 were alternatives.  It only needed to be caught by one in order for its supply to be standard rated.  It is very difficult to see how even the salads did not fall foul of condition (c):

has been kept hot after being heated

93.           The evidence was clearly that the meat, rice, beans and faitas were kept hot after being heated.

94.           Its case is that a purposive reading of the legislation ought to apply because (as I understand its case) people understand that salad is a cold food and Parliament wouldn’t have intended a salad to be treated as hot takeaway food, even though, as the appellant accepts, some of the items in its salads have been kept hot after being heated.

95.           I do not accept this submission.  If a salad by definition does not include any hot ingredients, then what the appellant provided was not a salad, whatever name it gave to it; if a salad by definition can include hot ingredients, then what the appellant provided was a salad but then there is nothing in its case that a purposive interpretation of the legislation would exclude salads as they are always cold.

96.           I find that a significant element of the salads sold by the appellant post 1 October 2012 included ingredients which had been kept hot after being heated.  Its appeal relating to the salads post September 2012 must be dismissed.

Conclusion

97.           I have concluded the liability issue entirely against the appellant:  it had in any event largely conceded its liability in the period post September 2012. 

98.           There is one caveat to this.  The evidence was that it was possible for a customer to order one of the five products on sale and only choose cold fillings.  While this may not have happened often, in so far as the appellant can show that it did happen, the product would have been zero rated as no part of it would have been above ambient air temperature at time of sale.  This is true before and after 1 October 2012.  It is possible, however, that this happens so infrequently that it is de minimis.

99.           Subject to this caveat on completely cold products, this is a decision in principle which has therefore finally determined the appeal in so far as it was against HMRC’s refusal to refund claimed overpaid input tax. 

100.       In so far as it was an appeal against assessments, issues of quantum also remain outstanding.  The parties should aim to settle the issues of quantum including the above caveat: if they are unable to do so, either party is at liberty to revert to the tribunal at any time for directions for a hearing to determine quantum.  Both parties should notify the Tribunal if and when they have settled the outstanding issues so that the file can be closed.

 

101.       This document contains full findings of fact and reasons for the preliminary decision. Any party dissatisfied with this preliminary 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. However, either party may apply for the 56 days to run instead from the date of the decision that disposes of all issues in the proceedings, but such an application should be made as soon as possible. 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: 6 MAY 2016