England and Wales Court of Appeal (Civil Division) Decisions

IN THE COURT OF APPEAL (CIVIL DIVISION)
ON APPEAL FROM THE UPPER TRIBUNAL
(Tax and Chancery Chamber)
Judge Greg Sinfield and Judge Jonathan Richards
[2019] UKUT 90 (TCC)

		Royal Courts of Justice Strand, London, WC2A 2LL
		20/07/2020

Covid-19 Protocol: This judgment was handed down remotely by circulation to the parties' representatives by email, release to BAILII and publication on the Courts and Tribunals Judiciary website. The date and time for hand-down is deemed to be 11.00 a.m. on Monday 20 July 2020

Lady Justice Asplin:

1. These appeals are concerned with the classification of the Vauxhall Vivaro (the "Vivaro") and the first and second generations of VW Transporter T5 Kombi van (the "Kombi" or "Kombis") for the purposes of income tax and national insurance contributions. The vehicles were provided by Coca-Cola European Partners Great Britain Limited ("Coca-Cola") to its employees for use in their work and for their own private purposes. The cash equivalent of the benefit of having been provided with a vehicle is treated as earnings from employment and taxed accordingly and there is a corresponding charge in respect of national insurance.
2. The classification of the vehicles is of importance because the cash equivalent of the benefit of a "van" and a "car", both of which are defined statutory terms, are calculated differently. Although the calculation is based on numerous factors, including the vehicle's emissions, in general, if a vehicle falls within the definition of a "car" the benefit and, as a result, the tax levied, will be greater than if it were a "van".
3. In fact, the question at the heart of both appeals is whether the vehicles in question were "goods vehicles", which is also a defined term. The question is of considerable importance. Large numbers of employees are supplied with Kombis or Vivaros, or vehicles which share their attributes.

Background

4. At the relevant times, Mr Payne and Mr Garbett were employed by Coca-Cola as technicians. In the tax year 2016-17 they were provided with a second-generation VW Transporter T5 Kombi van (a "Kombi 2"). In addition, in the tax year 2011-12, Coca-Cola provided other employees with the use of first-generation VW Transporter T5 Kombi vans (each a "Kombi 1") and Vivaros.
5. The Commissioners for Her Majesty's Revenue and Customs ("HMRC") determined that none of the vehicles were "goods vehicles" for the purposes of section 115(2) ITEPA 2003 and, therefore, were not "vans" and should be classified as "cars". They issued PAYE coding notices for 2016-7 to Messrs Garbett and Payne accordingly, and on the same basis, decided that Coca-Cola was liable for Class 1A national insurance contributions in the tax year 2011-12.
6. Messrs Garbett and Payne and Coca-Cola appealed to the First Tier Tribunal (Tax Chamber) (the "FTT") which heard the three appeals together. The parties asked the FTT to determine, as a point of principle, whether or not the Kombi 1, the Kombi 2 and the Vivaro were "goods vehicles". It seems that it was accepted that if they were not, they would fall within the definition of "cars" in section 115(1) ITEPA 2003 and the cash equivalent of the benefit would be calculated accordingly. In a decision dated 30 August 2017, the FTT decided that the Vivaro is a "goods vehicle" for the purposes of section 115(2) ITEPA 2003 but that both the Kombi 1 and the Kombi 2 are not: Payne and Others v HMRC [2017] UKFTT 655 (TC) (the "FTT Decision").
7. Coca-Cola and Messrs Payne and Garbett (together referred to as the "Taxpayers") appealed the FTT Decision that the Kombi 1 and Kombi 2 are not "goods vehicles" and HMRC appealed the FTT Decision that the Vivaro is a "goods vehicle" to the Upper Tribunal (Tax Chamber) (the "UT"). The UT upheld the FTT's decision: Payne & Ors v HMRC [2019] UKUT 90 (the "UT Decision"). The Taxpayers now appeal the UT Decision in relation to the Kombis and HMRC appeals the decision in relation to the Vivaro once more.

Relevant Statutory Provisions

8. In order to understand the UT Decision and the grounds of appeal and points raised in the Respondent's Notices, it is important to have the relevant legislation in mind. ITEPA 2003 contains provisions relating to income tax on employment income, amongst other things. Section 6(1)(a) imposes a tax on "general earnings", which is defined to include "earnings" and "any amount treated as earnings" (section 7(3)). Taxable benefits, set out in "the benefits code", are treated as earnings (section 7(5); section 63).
9. The benefits code dealing with cars, vans and fuel is at Part 3, Chapter 6 ITEPA 2003. It applies where a car or van is made available to an employee by reason of employment and is available for the employee's private use (section 114(1)). The "cash equivalent" of the benefit of the car or van is to be treated as earnings (section 114(2) (a) and (c)) and is calculated differently depending upon whether the vehicle is a car or a van (sections 120 and 121; and sections 154 and 155). There is also a corresponding charge to Class 1A NICs in s.10(1) of the Social Security Contributions and Benefits Act 1992.
10. The provision with which these appeals are directly concerned is section 115 ITEPA 2003. So far as is relevant, it provides as follows:

"115 Meaning of "car" and "van"

(1) In this Chapter—

"car" means a mechanically propelled road vehicle which is not—

(a) a goods vehicle,

(b) a motor cycle,

(c) an invalid carriage, or

(d) a vehicle of a type not commonly used as a private vehicle and unsuitable to be so used;

"van" means a mechanically propelled road vehicle which—

(a) is a goods vehicle, and

(b) has a design weight not exceeding 3,500 kilograms,

and which is not a motor cycle.

(2) For the purposes of subsection (1)—

. . .

"goods vehicle" means a vehicle of a construction primarily suited for the conveyance of goods or burden of any description;. . ."

As I have already mentioned, the FTT and the UT were asked to decide whether the Kombis and the Vivaro are "goods vehicles" for the purposes of section 115(2).

The Decisions Below

The FTT Decision

11. As the criticisms of the UT Decision inevitably relate to its treatment of the FTT Decision and turn, in some cases, on the precise language used, it is necessary to set out both decisions in some detail.
12. Having heard evidence, the FTT made detailed findings of fact about the characteristics of the Vivaro, the Kombi 1 and the Kombi 2 respectively. They are set out at [12] – [47] of the FTT Decision. The detailed characteristics of a Vivaro are at [14] – [23] and those of the Kombis are at [24] – [39]. General findings in relation to both types of vehicle are at [40] – [47]. Those findings were helpfully summarised by the UT at [7] of the UT Decision. They provide a useful background to the submissions made to us. The summary is as follows:

"(1) The Kombi 1, the Kombi 2 and the Vivaro as used by Coca-Cola's employees at the relevant times were all based on commercially available panel vans. Coca-Cola paid a third-party specialist contractor to modify those vehicles to make them suitable for their employees' use.

(2) The commercially available version of the Vivaro was effectively divided into just two sections: a driver and passenger seat at the front, with a relatively large storage area behind. The modifications made to the Vivaro included, but were not limited to, the addition of a second row of seats (that could accommodate two passengers). A window was added next to those seats and a steel bulkhead added behind those seats so that goods being transported could not enter the passenger compartment in the event of sudden braking (see [16(2)] and [16(7)]).

(3) Thus, Coca-Cola's modifications to the Vivaro resulted in the creation of a "mid-section" (of volume around 2.5 m3) including the additional two passenger seats that was separated from the rear cargo area by the steel bulkhead. The seats in the mid-section could be removed, but only with the use of tools. Even with the second row of seats in place, there would be around 1.5 m3 of space in the mid-section that could be used to carry goods (see [19] and [21]) and therefore the FTT concluded at [150] that the midsection of the Vivaro was adapted for the carrying of a significant amount of cargo. A number of other modifications were made to the Vivaro.

(4) The commercially available version of the Kombi 1 already included, as standard, a bench of seats that could seat up to three passengers behind the driver and single passenger seat at the front of the vehicle. This second row of seats was fixed to tracking on the floor of the Kombi 1 but could be removed without any tools. The commercially available Kombi 1 had windows on either side of this second row of seats.

(5) Coca-Cola's modifications to the Kombi 1 included adding a central partition behind this second row of seats to separate passengers from the rear load area and to prevent loose items entering the passenger compartment if the vehicle braked suddenly (see [26(2)]). Therefore, like the Vivaro, the Kombi 1 had a "mid-section" that included seating for passengers, although unlike in the Vivaro, this seating could be removed without tools. Coca-Cola added storage and racking to the rear section of the Kombi 1.

(6) The commercially available Kombi 2 had fundamentally the same design as the Kombi 1 although its front section included two seats for passengers other than the driver rather than one. Coca-Cola's modifications to Kombi 2 were similar to those of Kombi 1 and included the addition of removable racking in the mid-section that was suitable for the transport of goods (see [34(4)]). In relation to the Kombi 2, it was a contractual requirement imposed by Coca-Cola that the employee/driver of the vehicle had the racking in the mid-section in place during working hours. In other words, during working hours the second row of seats was required to be removed (see [39])."

13. The FTT considered the relevant principles to be applied at [127] – [144]. In summary, the FTT held that:

i) the word "construction" in section 115(2) ITEPA 2003 does not carry "any necessary reference to the 'structure' of an object in the sense of its core framework or its chassis or body"; [132]

ii) the construction of a motor vehicle comprehends the whole of its parts which allow it to perform its functions, even if those parts are removable [133];

iii) there is no statutory justification for the requirement that an alteration or modification must be fundamental in order to form part of the vehicle's construction and that the approach of the FTT in Timothy Jones v HMRC [2012] UKFTT 265 should be rejected: [134];

iv) construction was a wider concept than looking at the vehicle as it "rolled off the manufacturer's assembly line" and that its construction could change as the vehicle is modified or adapted and that the adaptation or modification does not have to be permanent to be relevant for the purposes of s115(2): [135];

v) the definition of "goods vehicle" requires one to determine whether the construction of a vehicle is primarily suited to the conveyance of goods or burdens and that in the case of 'multi-purpose' vehicles it is necessary to identify which 'suitability' is predominant: ([143]);

and

vi) it is necessary to take into account all of the characteristics of the vehicle, viewed objectively and not the particular use to which the employee puts the vehicle: ([144]).

14. The FTT then applied these principles to each of the three vehicles in question. In relation to the Vivaro, it noted that the commercially available version (prior to Coca-Cola's modifications) had the characteristics of a goods vehicle. In particular: the Vivaro's engine and transmission are mounted transversely and the driver's position set high to maximise the load area and load volume; the mechanical components of the Vivaro were packaged to allow a large flat load space and its height was designed to maximise the load and load volume; the design and dimension of the sliding door facilitated loading; and the suspension and braking system were designed to deal with heavy loads and a trailer: ([148]).
15. The FTT went on to consider the adaptations that were made to the Vivaro at [150] and [151]. As the significance of the adaptations in the mid-section of the Vivaro have featured in submissions, I shall set out [150] in full:

"150. In addition to the features listed in [148], the mid-section of the Vivaro was adapted to carry a significant amount of cargo, both behind the twin seats and to their left-hand side. This seemed to me to be an important feature in the overall assessment of the characteristics of the Vivaro. Clearly the majority of the mid-section was taken up by seating (which I recognise could only be removed with tools), but there was a material amount of cargo-carrying space (1.5m3 in Mr Roberts' estimation) which, in my view, could not be ignored . . . . This mid-section cargo-carrying capability which existed even when the mid-section seats were in the vehicle), when taken together with the rear cargo area, suggested to me that the primarily suitability of the Vivaro was for the conveyance of goods."

16. Having noted that despite the presence of windows in the mid-section, the level of comfort for passengers would have been relatively basic but that it was not a major factor in the overall assessment ([151]), the FTT concluded at [152] that:

"It is clear that the Vivaro had a dual capability of carrying passengers and carrying cargo. However, for the reasons I have given and taking account of all the characteristics of the vehicle, it seemed to me that, on a narrow balance, the construction of the Vivaro was primarily suited to the conveyance of goods."

17. The FTT then carried out the same exercise in relation to the Kombi 1 and Kombi 2. It noted that the Kombis shared many of the features of the Vivaro, referred to at [148] but that those basic design characteristics were also found in other Volkswagen vans such as the Shuttle and the Multivan, which were "essentially minibuses designed to carry passengers", and that those vehicles "were of a construction primarily suited to the carriage of passengers, even if the original design characteristics may have made them somewhat "over-engineered" for that purpose." ([156]).
18. The FTT concluded that the adaptations made to the Kombis did not make them more suitable for the carriage of goods; that the Kombi 1 and 2 were both equally suitable for carrying goods and passengers; and therefore did not constitute goods vehicles. In particular:

"161. In the case of the Kombi 1, therefore, the front row was primarily suitable for carrying passengers (including the driver), the mid-section was equally suitable for carrying passengers or (with the seats removed) goods. The rear cargo section was plainly primarily suitable for the conveyance of goods. In my view, therefore, it was not possible when looking at the vehicle as a whole to conclude that it was primarily suitable for the conveyance of goods. Looking at the entirety of the vehicle and taking all of its characteristics into account, it seemed to me equally suitable for carrying goods and passengers and cannot, therefore, be regarded as a "goods vehicle".

The FTT made clear that these conclusions also applied to the Kombi 2: ([162]).

The UT Decision

19. The UT considered the correct approach to the statutory definition of "goods vehicle" at [19] – [42] of the UT Decision. First, it determined that "a consideration of what vehicles might commonly be understood to be "cars" or "vans" is not directly relevant . . . " ([23]). The UT went on:

"[23] . . . Of course, in saying this, we are not deciding that a court or tribunal should apply the statutory definitions in a vacuum without regard to reality. It will be necessary to pay close attention to the construction of the vehicle and decide the particular use (if any) for which it is primarily suited. That exercise will often involve a consideration of the uses for which vehicles of a similar nature are suitable. However, we reject the submission that, simply because a vehicle answers to the description of a "van" as that term might be commonly understood, it necessarily follows that it is a "van", or a "goods vehicle", for the purposes of s 115 of ITEPA."

20. It turned, secondly, to the test of whether a vehicle is "primarily suited" to the carriage of goods for the purposes of the definition of "goods vehicle" in section 115(2) ITEPA. The UT decided that it was necessary to determine whether a vehicle had any primary suitability at all, and, if it had no primary suitability, it could not be a goods vehicle. See [24(1)] and [26]. It went on to conclude that "if a vehicle is of a construction marginally more suitable for the conveyance of goods than it is for any other use, its "primary suitability" is that of conveying goods" ([27]).
21. Turning to the use of the phrase "a vehicle of a construction primarily suited . . ." (emphasis added), the UT held that there is a link between "construction" and "structure" but that the link did not compel the conclusion that parts of a vehicle which are removable cannot be part of its structure and, therefore, are not relevant to whether the vehicle is of a construction primarily suited for the conveyance of goods. In particular, it held that the FTT had been entitled and correct to conclude that the removable seats in the Kombis were part of their construction and, therefore, relevant to whether they were within the goods vehicle definition ([31] and [32]).
22. At [33] – [42], the UT moved on to consider the significance or otherwise of the seats for passenger and driver at the front of the vehicles. This aspect of the UT and FTT's reasoning is the subject of detailed criticism by Mr Gardiner on behalf of the Taxpayers. The UT rejected Mr Gardiner's submission that seating for passengers and driver at the front of a vehicle could not, on their own, be indicative of a suitability for passenger transport but rather, all other relevant aspects of the vehicles should first be identified and, if those characteristics indicated a primary suitability for the conveyance of goods or burden, then the seating at the front should be regarded as ancillary to that primary suitability and "take its colour" from those other features. The UT saw no warrant for such an approach "in the context of a statutory provision that is clearly focused on identifying the "primary suitability" of a vehicle (if one exists) by reference to all relevant aspects of its construction" ([39] and [41]).
23. As to HMRC's appeal in relation to the Vivaro the UT held that there was no error of law in the FTT's conclusion at [152] that "on a narrow balance" the construction of the Vivaro was primarily suited to the conveyance of goods. It held that "[T]he FTT found that the Vivaro had two "suitabilities": carrying goods and carrying passengers. Once the FTT concluded that the construction of the Vivaro made it more suitable for carrying goods than passengers, if only by a fine margin, it followed that it was of a construction primarily suited to the conveyance of goods" ([44]). Secondly, the UT rejected HMRC's "Edwards v Bairstow" challenges to the FTT's factual conclusion relating to the mid-section of the Vivaro at [150] ([45] – [54]). Thirdly, it rejected the criticisms of the way in which the FTT had approached the expert evidence of Mr Roberts as to the features which were more characteristic of goods vehicles than cars. It found that the FTT had carried out a measured and balanced analysis of relevant factors at [147] and [148] of the FTT Decision ([57]).
24. As to the appeal in relation to the classification of the Kombis, the UT reiterated that the concept of "car" and "van" in ordinary parlance was of little relevance to the statutory definitions to be applied. It also held that the FTT had to perform a multi-factorial evaluation of whether the Kombis were of a construction primarily suited for the conveyance of goods or burden and that, overall, it did step back appropriately and consider its conclusions in the light of the overall characteristics of the vehicle, demonstrated by the final sentence of [161] which is set out at [18] above ([62]).
25. Further, the UT held that the FTT made no error of law in taking into account the presence of the Kombis' removable seating when evaluating whether they had a primary suitability and, if so, what for, nor did it consider that the FTT focussed on "use" ([64] and [65]).
26. The UT went on to the criticisms of the treatment of the front section of the Kombis at [69] – [75]. It held that the FTT had made an error of law in its evaluation of the front section of the Kombis "as the presence of a seat for the driver could not point in favour of the Kombis having a suitability for passenger transport since every road vehicle needs a driver". It did not accept the submission, however, that the presence of seating for passengers in the front section either pointed towards suitability for the carriage of goods or was a neutral indication ([71]). Further, it did not accept, as a matter of law, that the presence of seating in the front should "take its colour" from the features behind and was incapable of influencing the classification exercise ([72]). It was a matter for the FTT to evaluate what conclusion it drew from the presence of passenger seating at the front and there was nothing to suggest that there had been evidence before the FTT that, given the transportation of loads of the kind that the Kombis are suitable to carry, the presence of one or more passengers to help with loading and unloading, in addition to the driver, would be required, and that that evidence was ignored ([73]).
27. The UT also addressed criticisms of the FTT's conclusion in relation to the rear section of the Kombi and the FTT's ultimate conclusion in the following way:

"74. Mr Gardiner was also critical of the FTT's conclusion in [161] that the rear section of the Kombis was "primarily" suitable for the conveyance of goods. The FTT should, he submitted, have concluded that the rear section was solely suitable for that purpose. However, we do not think that the FTT's conclusion was to any extent influenced by a perception that the rear section of the Kombis was of a construction suitable to carry passengers or their effects. At [29], the FTT had found as a fact that the rear section of the Kombi could only be used for the purpose of carrying goods. Therefore, the conclusion in [161] that the rear section was "primarily" suitable for the carriage of goods is a typographical or drafting error rather than indicating an error in reasoning.

75. In conclusion, even though the FTT was wrong to draw the conclusion it did from the presence of a seat for the driver, overall it was entitled to reach the conclusion that the accommodation of seating for passengers in the front section pointed against the construction of the Kombis being primarily suitable for the carriage of goods or burden. Furthermore we consider that, after it had weighed the significance of its conclusion on the seating at the front against other competing considerations, it was open to the FTT to conclude that the Kombis had no overall "primary suitability" with the result that they were not goods vehicles. We therefore reject the criticisms of the Decision under this heading."

Grounds of Appeal and Respondents' Notices

Appeal No: 1203 - Kombis

28. The Taxpayers appeal the UT Decision in relation to Kombis on two grounds. First, the Taxpayers contend that the UT failed to apply the test imposed by statute and based its decision on the use of the interior of the vehicles rather than on their construction as a whole. This ground has three main strands. In particular, it is said that the UT: (a) failed to take account of the obvious inference to be drawn from the apparent structure of the Kombis; (b) looked at the use of the vehicles, and not at their construction as required by the statute; and (c) failed properly to take into account the terms "car" and "van" as an aid to the correct construction and application of the relevant definitions.
29. The Taxpayers' second ground of appeal is that further or alternatively, if the UT was right to adopt the same approach as the FTT in examining the internal disposition and use of the vehicles while placing little significance on the structure as a whole, which is denied, it fell into error in applying that approach by reaching a conclusion predicated on errors of law and fact, failing to consider Keeble v Miller [1951] 1 KB 601 and misconstruing the FTT's reasoning at paragraph 161 of the FTT Decision.
30. HMRC seek to uphold the UT Decision that Kombis are not "goods vehicles" on the basis of three additional reasons to those given by the UT. They are: (1) that "primarily suited" imports a requirement that suitability be more than marginal; (2) the UT was wrong to hold that the FTT had erred in concluding that the front row of a Kombi 1 was primarily suitable for carrying passengers (including the driver) because the presence of a seat for the driver could not point to suitability for passenger transport; and (3) that the UT was wrong to find that the FTT had made a typographical or drafting error at [161] of the FTT Decision when stating that "the rear section [of the Kombis] was "primarily suitable" for the carriage of goods".

Appeal No: 1215 - Vivaros

31. HMRC appeals the UT Decision in relation to the Vivaro on three grounds. They are that: (1) the UT erred in its construction of "primarily suited" in section 115(2) ITEPA 2003; (2) the UT was wrong to conclude that the FTT could lawfully conclude that the Vivaro had a primary suitability for the conveyance of goods, the only reasonable and lawful conclusion being that it was equally suitable for the conveyance of passengers and goods and did not have a primary suitability for the conveyance of goods; and (3) the UT was incorrect to find that the FTT's application of a "typicality" or "more characteristic" test did not lead the FTT into legal error.
32. In turn, the Taxpayers seek to uphold the UT Decision in relation to the Kombis on additional grounds. They are: (1) that the apparent structure of a Vivaro is that of a goods vehicle; (2) by virtue of its construction, the Vivaro is a goods vehicle; and (3) the front section of a Vivaro is of a construction primarily suited for the conveyance of goods or burden.

Discussion and Conclusions

33. Both Mr Gardiner QC on behalf of the Taxpayers and Ms McCarthy QC for HMRC agree that the classification of the Kombis and the Vivaro for the purposes of section 115 ITEPA 2003 should be the same and should be straightforward to determine. Their conclusions, however, differ. Although they both made detailed submissions in relation to the physical differences between the vehicles and the factual findings made by the FTT, their criticisms of the FTT and UT turn, for the most part, upon the Tribunals' construction of section 115 ITEPA 2003 and the application of the test as to whether a particular vehicle is a "goods vehicle" for the purposes of section 115(2). Accordingly, it is an appropriate place to begin.

Significance of the use of "car and "van"

34. Mr Gardiner began by making a general point about the construction of section 115 and, in particular, about the use of the terms "car" and "van" in section 115(1) which, he submits, should inform the interpretation of the section as a whole and the meaning of "goods vehicle" in section 115(2), in particular. He says that if one looks at section 115 as a whole, taking into account the juxtaposition of the definitions of "car" and "van" and the terms of section 115(1)(d), which excludes a vehicle "of a type not commonly used as a private vehicle and unsuitable to be so used" from the definition of a "car", a clear distinction is being made in the legislation between goods vehicles or trade vehicles as he described them, on the one hand, and vehicles which are for private use on the other. He says that it would be odd if a vehicle which one would ordinarily consider to be a van ended up falling within the definition of "car". He says that that is exactly what has happened in relation to the Kombis. In this regard, he took us to the photographs of the vehicles, noted the Coca-Cola logo on the Kombis and submitted that they look like vans.
35. In support of his submission, he took us to dicta from the decision of Megarry J in Roberts v Granada TV Rental Ltd S&U Stores Ltd v Gordon [1970] 1 WLR 889 which was concerned with investment allowances under section 16 of the Finance Act 1954. Under a proviso in section 16(3) the allowances could be made only if the vehicles were "of a type not commonly used as private vehicles and unsuitable to be so used." Megarry J concluded at 904F, that a distinction was being made "between private vehicles and trade vehicles, so that "private" is used in the sense of domestic, pleasure or social purposes."
36. Mr Gardiner also took us to the decision of Millett J in Gurney (Inspector of Taxes) v Richards [1989] 1 WLR 1180 which was concerned with whether a car with a blue flashing light fixed to the roof provided to a deputy chief fire officer was a "car" for the purposes of section 64(1) Finance Act 1976. A car for those purposes was defined by section 72(5)(a)(ii) as excluding a vehicle "not commonly used as a private vehicle and unsuitable to be so used". Megarry J's approach to "private" in the Granada case was adopted and it was decided that the vehicle in question fell within the exclusion in section 72(5) because it was illegal for a member of the public to drive it on the road.
37. It seems to me that whilst Mr Gardiner's approach to section 115 appears to be straightforward, the distinction he seeks to make between trade vehicles and private vehicles is unhelpful. It is clear both from the wording used in section 115 and the context as a whole that that was not the dichotomy which the statutory draughtsman sought to create. The provision must be viewed in context and construed as a whole, as it stands. The section arises in the context of determining the cash equivalent of the benefit of having a vehicle made available for private use whether it is a "car" or a "van". Although it is true that the private use of a car is likely to be greater than that of a van, the very nature of the cash equivalent provisions assume that either type of vehicle may be used privately and it is the cash equivalent of that benefit which is to be taxed, save in the case of a van in a tax year in which the private use by the employee, or member of his family or household, is insignificant: (section 114(3A)). Once a vehicle falls within one of the definitions, the benefit will be calculated in accordance with its environmental impact amongst other things. It seems to me, therefore, that Mr Gardiner's approach would undermine the statutory definitions which were chosen and the purpose of the legislation itself.
38. His approach also places too much weight upon the phrase "not commonly used as a private vehicle" in section 115(1)(d). Each of the exceptions from the definition of a car in sub-sub-sections (a) – (d) are disjunctive. It seems to me, therefore, that the reference to "private vehicle" in (d) cannot colour the meaning of "car" or "goods vehicle" and create the juxtaposition which Mr Gardiner suggests. It is merely a means of defining a category of vehicle which must be excluded. Accordingly, a car with dual controls used by a driving instructor or the car in the Gurney case with a blue flashing light on the roof which it was illegal for a member of the public to drive on the road, would not be "cars" as a result of section 115(1)(d).
39. In any event, it seems to me that neither the Granada nor the Gurney case provides support for Mr Gardiner's approach. They were concerned with statutory provisions which arose in different contexts and which made the precise distinction which Mr Gardiner seeks to impose on very different wording here.
40. Furthermore, in my judgment, the effect of Mr Gardiner's dichotomy between private and trade vehicles is to override or undermine the definition of "goods vehicle" and cannot be correct. One is required to construe and apply that definition. If it is satisfied, and the remainder of the definition of a "van" in section 115(1) is also fulfilled, the vehicle will be classified as a "van". If not, it is possible, but not inevitable, that the vehicle may fall within the definition of a "car".
41. It seems to me, therefore, that nothing can be gained from the use of "car" and "van" in section 115 beyond the fact that they are the chosen defined terms. It follows, therefore, that I do not consider that the UT was wrong in its approach to those definitions at [22] and [23] of the UT Decision. In any event, as Ms McCarthy pointed out, the UT, and the FTT before it, was concerned with the statutory definition of "goods vehicle" and were not directly concerned with "van" or "car".

The meaning of section 115(2) - "construction"

42. How should the definition of "goods vehicle" be construed? First, what is meant by a vehicle "of a construction primarily suited for the conveyance of goods" (emphasis added)? It is not in dispute that the actual use of a vehicle is not relevant to the question of whether a vehicle is a goods vehicle for the purposes of section 115: C & E Commissioners v Jeynes [1984] STC 30 at [31j] and Flower Freight Co. Ltd. v Hammond [1963] 1 QB 275 at [283].
43. Mr Gardiner submits, however, that the appearance or apparent structure of a vehicle is of particular importance. He says that the starting point is to look at the vehicle and determine what type it is. In this case, he says that the answer is obvious. I agree, however, with the UT that the fact that a vehicle may look like a van is not conclusive. Appearance, or apparent structure, as it is termed in the grounds of appeal, albeit a factor to be taken into consideration when applying the statutory test, is not the test itself. Nor is the fact that some vehicles bear company logos. It seems to me that that is irrelevant. The presence of a logo does not go to the question of whether a vehicle is "of a construction primarily suited for the conveyance of goods or burden . . ." (emphasis added).
44. Whilst Mr Gardiner's emphasis upon appearance may seem superficially to have the attraction of certainty, such an approach impermissibly places too much emphasis on appearance and external features over suitability and undermines the use of the word "construction" in the statutory definition of "goods vehicle" and the definition itself.
45. Furthermore, it seems to me that Mr Gardiner asks the passage in Keeble v Miller [1950] 1 KB 601, to which he referred in support of his approach, to bear too much weight. In that case, Lynskey J, with whom Lord Goddard CJ agreed, stated at 605 that in order to decide whether a vehicle was "constructed" for the purpose in section 2(1) Road Traffic Act 1930, ". . . the magistrate ought to see the vehicle and have its dimensions put before him, and then ask himself the question: is that vehicle in that condition constructed to carry a load or passengers, apart from the driver and his mate?" It seems to me that Lynskey J was not advocating an approach based solely upon the appearance or apparent structure of a vehicle. He was requiring the decision maker to take account of all aspects of the "condition" of the vehicle.
46. The same is true of Mr Gardiner's reliance upon the passage in the judgment of Lord Parker CJ in Flower Freight case. That case was concerned with whether a vehicle had been " . . . constructed or adapted for use for the carriage of goods" for the purposes of the section 191 Road Traffic Act 1960. A substantial roof rack had been placed on a passenger vehicle suitable for the carriage of thirteen passengers and goods were carried on it. The defendants were charged with using a goods vehicle on a road for the carriage of goods for hire or reward without the appropriate licence. Lord Parker CJ held at 282-3:

"Accordingly, "constructed . . . for use for the carriage of "goods" must mean something more than constructed so as to be capable of carrying goods. Nor do we think that the intention with which a particular vehicle is constructed or adapted can be a relevant consideration. The use to which a particular vehicle has been put is, in our view, as irrelevant as the intention with which it was bought or altered. A limousine does not become a goods vehicle because it carries potatoes and no passengers; nor does a lorry cease to be a goods vehicle because it carries passengers and not goods. The question is not what does this particular vehicle usually carry nor what is this vehicle capable of carrying, but what is the use for which the vehicle was constructed or adapted. It seems to us that this question falls to be resolved by looking [at] the vehicle and considering whether vehicles of this kind are ordinarily used for the carriage of passengers and their effects, or the carriage of goods. In this connection, it may well be that the manufacturers' advertisements, etc. can be looked at, not to ascertain the intention or purpose of the manufacturer or purchaser in relation to the particular vehicle, but as some evidence of the use to which vehicles of the same type are ordinarily put."

It seems to me that Lord Parker CJ was considering whether a vehicle had been "constructed or adapted . . ." by considering all of its features in order to determine its purpose or function. He was not advocating an approach based solely upon outward appearance.

47. Next, Mr Gardiner submits that only features or modifications which are fundamental to the structure of the vehicle are part of its "construction" for the purposes of the statutory definition of "goods vehicle". He says that the starting point here is the FTT's finding that the Kombi is based on a panel van: [24] and [25]. He says, therefore, that it is necessary to identify a feature or features which amount to a fundamental change to that original design as a panel van in order to justify the conclusion that it is not "of a construction primarily suited for the conveyance of goods" (emphasis added). He says that there is no such fundamental change to the structure or construction of the vehicles.
48. Ms McCarthy, on the other hand, submits that Mr Gardiner's incremental approach is wrong. She says that "construction" in the definition of "goods vehicle" means nothing more than the physical features of the vehicle as it stands at the time of the charge to tax and that one must assess the construction in the context of suitability for a purpose. She says that one should look at the purpose of that construction in terms of the use which it was intended to perform. What is the vehicle's functionality as it stands?
49. She also submits that Mr Gardiner is wrong to rely upon Coleborn (T) & Sons Ltd v Blond [1951] 1 KB 43 and the Timothy Jones case for the proposition that one should start from the premise that the Kombi is a panel van and seek to identify a fundamental change thereafter. In the case of Coleborn, the Court of Appeal was concerned with whether purchase tax was due pursuant to the Finance (No 2) Act 1940 and the Finance Act 1946, in relation to a vehicle which was first constructed with seats for an officer and driver in the cab and had field radio apparatus and two seats for those operating it, in the rear. It was converted into a shooting brake. The wheels, engine, bonnet and scuttle were the same as when the vehicle was originally construed, but it had eight seats, a pair of back doors, a permanent roof and a new floor. The question was whether the work done was a "process of manufacture which made, or was in the course of making, a vehicle constructed solely or mainly for the carriage of passengers": section 16(3) Finance Act 1946. Bucknill LJ made the following obiter observation at 48:

" . . . I think that the concluding words of s.16, sub-s. 1, of the Finance Act, 1946, indicate that even if the vehicle was originally constructed to carry passengers, yet, if the work done by the plaintiffs was to manufacture a thing which was different from what it was before the work was done, in other words, if the work was such as to involve an alteration in the article so fundamental as to produce a different article, then purchase tax would be chargeable on the article so produced. . . ."

50. In my judgment, the FTT was right to reject the approach in the Timothy Jones case and the UT was correct to decide as they did at [40] that Bucknill LJ's obiter remarks were made in a different statutory context which makes all the difference. The Court in that case was concerned to determine whether the work done was a process of manufacture which made the vehicle one which was constructed solely or mainly for the carriage of passengers (see section 16(1) Finance Act 1946) and it arose in the context of the alteration of an actual vehicle. As the court was concerned with whether there had been a "process of manufacture", it is easy to see that it was looking to see whether the changes had been "so fundamental as to produce a different article".
51. It seems to me that when construing and applying the definition of "goods vehicle", there is no reason to begin with an original design, if there is one, and to seek to identify fundamental changes to it in order to fall within the phrase "of a construction . . .". The fact that a manufacturer may have started with a basic design for a vehicle and adapted it for a variety of purposes to create different vehicles is not relevant to its construction. The construction is that of the final article. The construction cannot be different because one manufacturer starts from scratch and another uses an existing design for the basis of a new model or type of vehicle.
52. It seems to me, therefore, that both the FTT and UT were correct to decide that the approach of the FTT in the Timothy Jones case should be rejected for the reason the UT gives at [30] of the UT Decision. It is not necessary to look for fundamental alterations to an original structure in order to fall within the phrase "a construction primarily suited . . . " As the FTT decided at [132] of the FTT Decision, "construction" simply means the manner in which the vehicle has been put together, assembled or built. It seems to me, therefore, that the question of whether a vehicle falls within the definition of a "goods vehicle" requires one to look at its construction as a whole at the date at which the test is applied. There is no need to look into its antecedents, if it has any, and to work from there.
53. Such an approach is consistent with Keeble v Miller and the Flower Freight case although, as I have already mentioned, they were concerned with different legislation. In Keeble v Miller it was held that the relevant time at which the construction of the vehicle must be tested was that at which the alleged offence was committed. In the Flower Freight case, Lord Parker LC held at 284 - 5 that ". . . the words "original construction" and "structure" as used in the cases are wide enough to include the general "design" of the vehicle and should not be limited to its "framework"".
54. It follows, therefore, that in my judgment, one should consider the Kombis and the Vivaro for that matter, in their modified form, and should not start from the premise that they were based on panel vans for the conveyance of goods and look for sufficient alterations to justify moving away from that original function. The term "construction" cannot be taken to mean the construction of the vehicle as it rolled off the factory production line. Such an approach would be contrary to the purpose of Chapter 6, ITEPA 2003 which is to ascertain the taxable benefit that is in the hands of the employee in the tax year in question.
55. It also follows from the conclusion that "construction" of a vehicle is not confined to fundamental changes to its structure, that depending on the facts, fixings for removable seating and the seating itself may be part of the "construction" of a vehicle and that one is not looking for changes to an original structure.
56. In this regard, it seems to me that Taylor v Mead [1961] 1 WLR 435, to which Mr Gardiner took us, can be distinguished on a number of grounds. In that case, the defendant was charged with using a car for a purpose which rendered it liable to a higher rate of duty before such duty was paid, contrary to section 13(2) Vehicles (Excise) Act 1949 (the "1949 Act"). He was a commercial traveller and had erected a rail across the width of the rear of a saloon car and two cross rails, in order to hang dresses and other articles of clothing. The rails were attached to the door pillars and end of the car by screws and a bracket behind the rear seat. The justices considered that the car was not "constructed or adapted for use . . . for the conveyance of goods" within the meaning of the definition of goods vehicle in section 27(1) of the 1949 Act. Lord Parker CJ held as follows at 438:

" . . . It seems to me that, by the conjunction of the words "constructed or adapted," the definition is really saying "originally constructed or where the structure is subsequently altered." Immediately one says that, the question arises whether it can be said that the structure of the vehicle in the ordinary sense of the word has been altered, or whether the structure remains the same, but that some small fitting or attachment is made which, although it physically involves making small holes for screws in the structure, could not in any ordinary sense of the word be an alteration of the structure."

57. Lord Parker CJ was concerned with a provision which is substantially different from the definition of "goods vehicle" in section 115(2) ITEPA 2003. As he pointed out, as soon as one takes account of the phrase "constructed or adapted" one is concerned with whether an original structure has been altered. That does not follow from the mere use of "of a construction" in section 115(2). Furthermore, the facts in that case were very different. Small holes for screws are different from fixing for seating provided with the vehicle.
58. It seems to me, therefore, that the UT was right to decide at [32] of the UT Decision that the FTT was both entitled, and correct, to conclude that the Kombis' removable seats were part of their construction and relevant to the question of whether the Kombis were of a construction primarily suited to the conveyance of goods or burden.

The meaning of section 115(2) - "primarily suited"

59. What is the ordinary and natural meaning of the phrase "primarily suited" in the context of section 115(2) ITEPA as a whole? In the course of submissions, Mr Gardiner stated that in order to determine the meaning of "primarily suited" in any particular case, one should look at the design purposes of the vehicle and determine what it is suited to first. As I have already mentioned, he says that this is resolved by looking at what type of vehicle it is and determining whether it is a trade vehicle or a private vehicle for passenger use. He says that if one looks at the relevant features of a Kombi referred to at [24] and [25] of the FTT Decision they nearly all point to suitability for the carriage of goods and accordingly, this is not a marginal case at all.
60. Ms McCarthy on the other hand says that one needs to look at the features of a vehicle in the round in order to determine for what it is "suited" or appropriate for. In this regard, she relies upon Megarry J in the Granada case at 904H – 905A where he stated that ""Suitable" seems to me to bear the meaning of "fitted for, adapted or appropriate . . ." Ms McCarthy also emphasises that a vehicle may be a multi-purpose or a hybrid. She gave the example of an amphibious vehicle which has features of a land vehicle and a boat. She says that in relation to each feature it is necessary to consider whether, whilst it may be suited to one purpose, it actually detracts from the other. She submits, therefore, that "suited" in relation to each feature, does not mean "designed for" and gave, as an example, that the fact that strong brakes may be suited in the sense of designed for the conveyance of goods does not prevent those brakes from being suited for the carriage of passengers as well.
61. In fact, Ms McCarthy agreed with Mr Gardiner that whether a vehicle is "primarily" suited to a particular purpose is something which should be clearly discernible by the taxpayer and should not turn on fine distinctions. She accepted that the FTT was right to use "predominantly" as a synonym for "primarily" and submitted that the suitability should be more than incidental. Furthermore, she says that although the FTT was right to focus on predominance, it was wrong in the way in which it measured that predominance and indulged in too great a level of analysis. In order to be "primarily suited" the purpose must be more than marginal.
62. I agree. It seems to me that it is clear that the use of "primarily" imports something more than a suitability which is marginally greater than another or any number of other suitabilities. In my judgment, therefore, the UT was wrong to conclude at [27] that "if a vehicle is of a construction marginally more suitable for the conveyance of goods than it is for any other use, its "primary suitability" is that of conveying goods". "Primarily" requires more than a percentage point advantage of one suitability over another. To put the matter another way, in my judgment, the natural and ordinary meaning of "primarily" requires the decision-maker to determine what the vehicle in question is first and foremost suitable for. Its suitability must be considered in the round and is not merely the produce of a mechanical or mathematical exercise.
63. I also agree with Ms McCarthy, therefore, that it is possible, although not inevitable in every case, that a multi-purpose vehicle may not have a primary suitability at all. On the other hand, a vehicle may have numerous purposes, but, nevertheless, be primarily suited to one of them.
64. I also agree that "suited" can be equated with "appropriate" and that a vehicle may have features which are appropriate for more than one purpose. In such circumstances, as Ms McCarthy stated, it is possible for a feature to be appropriate for a particular purpose whilst being over-engineered for another, as the FTT observed at [156] of the FTT Decision in relation to the "Shuttle", a passenger vehicle, based upon an original panel van design. Such over-engineering does not of itself render the feature unsuited to another purpose. It is necessary to consider the matter in the round.
65. It also follows that I agree with Ms McCarthy that when seeking to carry out an evaluation, it is misleading to look at each feature of a vehicle separately and to determine whether that feature is more typically found in one type of vehicle or another. Not only is "typicality" not the relevant statutory test but also a feature may be typically found in a van because it is suitable for the conveyance of goods, such as strong brakes, but may not detract from another suitability. The presence of such a feature does not necessarily mean that the vehicle is "primarily suited" to one use or another. Accordingly, in my judgment the FTT was wrong to be swayed by whether features were "more characteristic" or "more typical" to a type of vehicle, as it seems to have been at [147] and [148] of the FTT Decision.

Conclusion in relation to the proper construction of the term "goods vehicle"

66. It follows that the Taxpayers' first ground of appeal in relation to the Kombis fails. In my judgment, neither the FTT nor the UT failed to construe section 115 and the definition of "goods vehicle" correctly in the way which Mr Gardiner suggested. Furthermore, neither the FTT nor the UT erred in their approach to the construction of the Kombis in the way which Mr Gardiner submitted they had. As I have already mentioned, the FTT was right to reject the approach adopted in the Timothy Jones case and to take account of the flexible layout of the Kombi and the removable seats.
67. Furthermore, HMRC's first and third grounds of appeal in the Vivaro appeal succeed. It is not sufficient, as both the FTT and the UT held, to conclude "on a narrow balance" or by a "fine margin" that a vehicle's construction is "primarily suited for the conveyance of goods" (emphasis added). Furthermore, the UT was wrong to hold that the FTT's application of a "typicality" test did not lead it into error.
68. Having arrived at these conclusions it remains to consider the remaining grounds of appeal.

The Kombi Appeal 1203 – Ground 2

69. The second ground in the Kombi appeal is concerned with errors of law and fact and what is said to be the UT's misunderstanding of the FTT's reasoning at [161] of the FTT Decision. Mr Gardiner submitted that on the findings of fact reached by the FTT, the only available conclusion, as a matter of law, was that the Kombis are goods vehicles. He says that the FTT and the UT were wrong about the treatment of the front row of the Kombi and the effect it had on the ultimate conclusion, were wrong about the mid-section, failed to take account of the finding that the rear section was "solely" for goods and that the FTT failed to stand back and look at the vehicle as a whole.
70. It is necessary, therefore, to consider the FTT's findings in more detail. First, Mr Gardiner looked at each of the features of the Kombi 1 enumerated and described by the FTT at [25] of the FTT Decision. The modifications made to the Kombi 1 by a third party specialist contractor were set out in six numbered sub-paragraphs at [26] of the FTT Decision. Mr Gardiner submitted that but for the features at [25] (10) and (11) all of the twelve features set out in that paragraph are for the carrying of loads. He accepted that the features at (10) and (11) do not directly facilitate the conveyance of goods. They are:

" (10) There was a three-seater bench seat with seat belts and headrests. The seat fitted into mounting locations in the floor and could be removed without tools. Therefore, the Kombi 1 could seat up to five passengers including the driver. Mr Sayer's evidence was that Coca-Cola chose Seat Pack option B for the front row which contained a single passenger seat together with the driver's seat.

(11) There were windows on each side of the second row of seats. These windows had an opening panel."

71. Of the six modifications, Mr Gardiner submitted that they were all for the carriage of goods save possibly for number (2) which was "a central partition behind the second row of seats to separate the driver and passengers from the load area, preventing loose items from entering the passenger compartment if the vehicle braked suddenly". Mr Gardiner submitted therefore, that, as far as the Kombi 1 was concerned, on the facts as found by the FTT, there was nothing in the modifications that detracts from the Kombi 1's suitability for the conveyance of goods.
72. He also submitted that the FTT's further findings at [27] – [30] do not alter the primary suitability of the construction of the Kombi. At [27] the adaptations are noted, at [28] reference is made to the removable seats in the mid-section which Mr Gardiner says are not relevant or at least, neutral and to the space for goods and at [29] the FTT recorded its finding that the rear section was only for the purpose of the carriage of goods.
73. He conducted the same exercise in relation to the Kombi 2. The FTT stated at [31] that the Kombi 2 was "fundamentally the same design as the Kombi 1 with certain minor updates to the specification and features". One of those updates was the inclusion of two rather than one passenger seat in addition to the driver's seat in the front: [32(12)]. Its characteristics were set out in twelve numbered sub-paragraphs at [32] of the FTT Decision. The modifications set out in [34] were similar to those made to the Kombi 1 and it was noted that the "adaptations could be removed, but all would have required permanent fixing holes or fittings in the vehicle structure".
74. The FTT made additional findings about the mid and rear sections as follows:

"36. Thus, the Kombi 2 was constructed by the manufacturer with a second removable second row of seats (in a 2+1 combination), and windows on both sides of that row. Following adaptations by a third party contractor, it had optional racking in the mid-section and fixed racking in the cargo area. When the racking was inserted in the mid-section, the second row of seats had to be removed (and vice-versa). It was, however, possible for the double seat, in the mid-section, to be removed and racking inserted on the right-hand side of the vehicle whilst retaining the single seat on the left-hand side of the vehicle.

37. With both of the storage units fitted in the mid-section, this area (of approximately 2.5 m³) could only be used for the conveyance of goods. The combined volume of the storage units in this area was approximately 0.8 m³. With neither the seats nor the storage units fitted to this area, the full volume of the area would be 2.5 m³ and could be used for the purpose of carrying goods.

38. The rear section of the vehicle (i.e. behind the bulkhead) was approximately 3 m³ and contained the racking and storage units. This area could only be used for the conveyance of goods.

39. In relation to the Kombi 2, it was a contractual requirement imposed by Coca-Cola that the employee/driver of the vehicle had the racking in the mid-section in place during working hours. In other words, during working hours the second row of seats was removed."

75. Once again, it was submitted that all of the features pointed to suitability for the conveyance of goods but for those at [32] (10) and (11) which were passenger seats in the second row which could be removed without tools and windows and doors in each side of the mid-section. Standing back, therefore, Mr Gardiner says that on the facts as found, both Kombis are primarily suited to the conveyance of goods, having been based on the panel van design and that there was nothing to displace that conclusion.
76. In this regard, Mr Gardiner also reminded us that despite having reached a different conclusion in relation to the Vivaros, the FTT noted at [154] that the Kombis had the same type of original design features as the Vivaro which had been set out at [148] of the FTT Decision. The FTT went on, nevertheless, to hold that those features were "relevant but they are not sufficient to satisfy the test in section 115(2) if other features of the vehicles exist which indicate that the suitability of the vehicles was not primarily that of conveying goods".
77. Mr Gardiner submits that there were no "other features" to which the FTT referred in the subsequent paragraphs and that the only proper conclusion was that the Kombis should be classified as goods vehicles in the same way as the Vivaro. He says, therefore, that the FTT's conclusions at [161] are wrong in a number of respects. His more detailed criticism of the UT's approach is most easily understood by reference to its conclusions and the FTT's findings in relation to each section of the Kombis.
78. Before approaching the FTT and UT's analysis in this way, as a matter of convenience, it is important to make clear that I do not accept Mr Gardiner's criticism of the last sentence of [161] of the FTT Decision and the UT conclusion in that regard at [70] of the UT Decision. In my judgment, the UT was correct to decide that although the FTT referred to the characteristics of different sections of the Kombi, it properly had regard to aspects of all three sections of the Kombi 1 and also stood back and considered the suitabilities of the vehicle as a whole.

Front section

79. First, Mr Gardiner criticised the FTT's approach to the front section of the Kombi which it concluded was "primarily suitable for carrying passengers (including the driver . . ." ([161]). He also criticised the UT's rejection of the argument that seating for passengers in the front section takes its colour from the suitability of the remainder of the vehicle and points either towards suitability for the conveyance of goods, or was neutral, and its conclusion that it was a matter for the FTT to evaluate, despite having decided that the FTT had been wrong about the significance of a seat for the driver: ([69] – [73] and [75] of the UT Decision).
80. Mr Gardiner made detailed submissions about the seating in the front section of the Kombis and again took us to Keeble v Miller and, in particular, to the passage in the judgment of Lynskey J at 605 at which the driver and his mate are excluded from the question of whether a vehicle is constructed to carry a load or passengers. He also referred us to Cook v Hobbs [1911] 1 KB 14 in which the question before the court was whether a purpose built trap was a "cart . . . which is construed or adapted for use, and is used, solely for the conveyance of any goods or burden in the course of trade or husbandry". The trap was used to take livestock and goods to market and the farmer took his wife and son with him to assist in loading and unloading and selling the goods. Lord Alverstone CJ held at 17 that "burden" included the persons who were taken to market to assist in loading, unloading and selling the goods which had been transported. The same approach was adopted by Bucknill LJ in the Coleborn case. He observed at 47 that every vehicle must have a driver and that "if carrying passengers is only incidental to the use of the vehicle for other purposes, it would be exempt" from section 16 Finance Act 1946 which was concerned with "a vehicle constructed solely or mainly for the carriage of passengers".
81. Mr Gardiner submitted, therefore, that one must look to the function of the vehicle and that the facility for the carriage of passengers in the front section does not detract from that function. A van needs a driver and a mate to assist with loading and unloading and they need somewhere to sit. Mr Gardiner submits, therefore, that the seating in the front section takes its colour from the rest of the vehicle. In this case, he says, the Kombi has a rear cargo section with a 3m³ capacity and a mid-section with a 2.5m³ capacity for goods and the front section takes its colour from them. He says therefore, that the rear and mid-sections are primarily suited for the conveyance of goods and, therefore, the front row is also suited in the same way.
82. He says, therefore, that the UT was wrong, at [71] and [72] of the UT Decision, having held that the FTT made an error of law in its evaluation of the front section of the Kombis because the presence of a seat for the driver could not point in favour of a suitability for passenger transport because every road vehicle needs a driver, nevertheless, to go on to reject his submission that the presence of seating for passengers takes its colour from the features of the remainder of the vehicle which were for the conveyance of goods and to hold that it was a matter for the FTT to evaluate on the evidence.
83. Although Ms McCarthy criticises Mr Gardiner's analysis of the cases to which we were referred, in the end she too accepts that the suitability of the front row of the Kombis takes its colour from the other sections. Her conclusion, however, is different. She says that there is no evidence that there was a driver's mate in these vehicles and that one should not ignore the fact that the Kombi is also suited to transport a crew of six on the basis of three seats in the mid-section and three in the front. She submits, therefore, that the Kombi has a flexible layout from which the front section takes its colour and therefore it is primarily suited to either goods or passengers.
84. It seems to me that the difference in approach to the front row of the Kombi is driven by the way in which Mr Gardiner and Ms McCarthy wish to characterise the vehicle as a whole. Although they both say that the front row takes its colour from the remainder the vehicle, they do so having characterised the remainder differently and, in Ms McCarthy's case, having also taken account of the alternative passenger layout which includes the front section.
85. I agree with the way in which the UT approached this matter at [71] – [73]. Although the presence of a driver's seat may be neutral, it was necessary for the FTT to evaluate what conclusions it drew from the possible configurations of the Kombi and the UT was right to decide that the FTT was entitled to come to the conclusion it did on the evidence before it.

Rear section

86. It is most convenient to deal with the rear section of the Kombis next. Although Mr Gardiner did not press the matter in oral submissions in his written argument he took the point that the FTT concluded that the rear section was "primarily" suited for the conveyance of goods at [161] but had already decided that it was "solely" suited to goods earlier in the FTT Decision. He criticised the UT conclusion at [74] of the UT Decision that the use of "primarily" by the FTT at [161] of the FTT Decision was merely a typographical error and did not indicate an error of reasoning.
87. It seems to me that the UT approach to the matter at [74] reveals no error of law. As the UT pointed out, there is nothing to suggest that the FTT's conclusion in relation to the rear section was to any extent influenced by a perception that that section was of a construction suitable to carry passengers or their effects and it had earlier found that it was solely for the conveyance of goods. Furthermore, despite Ms McCarthy's detailed submissions based on the decision in the Flowers case, to the effect that the rear racking and shelving was equally suitable for carrying goods or passenger effects, it seems to me that in the absence of an error of law, it is not for us to re-evaluate the suitability of the rear racking.

Mid-section

88. In relation to the mid-section of the Kombis, Mr Gardiner's starting point is that the FTT and the UT took the wrong approach to the removable seating because they misconstrued "construction" in section 115(2) ITEPA 2003. I have already decided that the approach in the Coleborn case is not relevant in this context and that the FTT was right to reject that approach in the context of section 115(2). I have also decided that Taylor v Mead can be distinguished from this case. It follows, therefore, that in my judgment, neither the FTT nor the UT erred in its approach to the removable seating. The FTT was entitled to find as it did at [158] – [160]. It concluded that both versions of the Kombi were supplied with seats which "when fitted took up most of the space in that section" but when removed, "almost the entire area of the mid-section was available for conveyance of goods" ([158]). Accordingly, the FTT decided that the mid-section was "equally suited for either purpose" ([160]) and that in essence, Kombis are "multi-purpose vehicles" ([159]).
89. As the UT pointed out at [32] of the UT Decision, the FTT had before it evidence that "indicated that the Kombis were manufactured and sold with the removable seating in place and that the manufacturer's brochure described the Kombis as "flexible, versatile and extremely adaptable . . . offering seating for up to five passengers" which it would have been wrong to ignore".

Conclusion

90. It seems to me, therefore, that Mr Gardiner's attack upon the FTT and UT's conclusions, based upon these further alleged errors of law and reasoning and upon the basis of Edwards v Bairstow, must also fail. In my judgment, the FTT was entitled to come to the conclusions it did based upon its findings of fact and to come to those findings based upon the evidence before it. The UT in its turn was entitled to hold as it did in relation to the Kombis.
91. Accordingly, for all the reasons set out above, I would dismiss the Taxpayers' appeal.

The Vivaro

92. As I have already mentioned, it also follows from my conclusions in relation to the proper construction of section 115 and the proper approach determining whether a vehicle is a "goods vehicle" that in my judgment, the FTT and the UT applied the wrong test in relation to the Vivaro. Having decided that "primarily" in section 115(2) envisages that a vehicle may have more than one potential suitability and that it "requires the taxpayer to demonstrate that the predominant suitability of the vehicle in question is for the conveyance of goods or burden" ([143] of the FTT Decision), the FTT went on to evaluate each feature of the Vivaro separately at [146] – [152] and having decided that the vehicle has a "dual capability", nevertheless, came to a conclusion "on a narrow balance" that it is primarily suited to the conveyance of goods. In doing so, it seems to have lost sight of the concept of predominance. The UT endorsed that erroneous approach at [44] of the UT Decision. As I have already decided, it is not enough to slip past the post. "Primarily" means something more than a suitability which is first in the list by a whisker. It means first and foremost. It cannot encompass very narrow margins. It is also possible for a multi-purpose vehicle to have no primary suitability at all. The UT and the FTT, therefore, proceeded on the basis of an error of law.
93. Furthermore, in deciding that the Vivaro was primarily suited for the conveyance of goods, it seems to me that the FTT was swayed by its view of the mid-section of the Vivaro which it described and evaluated at [150] of the FTT Decision which is set out at [15] above. The so-called difference in layout from that of the Kombi is the space for goods/tools in the Vivaro mid-section. In this regard, the FTT seems to have lost sight of the fact that it had found at [34(4)] that the mid-section of the Kombi 2 also had removable storage units. Even on the FTT's view that a narrow balance is enough to be "primarily suited", it seems to me that the difference is insufficient upon which to differentiate the Vivaro from the Kombi and to decide that the Vivaro is primarily suited for the conveyance of goods. This is all the more so when one considers whether the Vivaro was suited first and foremost for that purpose. It seems to me that the only reasonable conclusion on the facts, having weighed all the features of a Vivaro and considered the vehicle as a whole (including its appearance) is that it, like the Kombis, is multi-purpose and is not primarily suited to the conveyance of goods.
94. Accordingly, for all of the reasons set out above, I would allow HMRC's appeal.

Lord Justice David Richards:

95. I agree.

Lord Justice Patten:

96. I also agree.

---

ORDER

---

UPON HEARING Mr John Gardiner QC and Mr Ed Hellier on behalf of the Taxpayers and Ms Hui Ling McCarthy QC and Mr Hugh Flanagan on behalf of HMRC

IT IS ORDERED THAT:

The Taxpayers' appeal (reference A3/2019/1203) is dismissed.

HMRC's appeal (reference A3/2019/1215) is allowed.

The decision of the First-Tier Tribunal is varied such that the appeal of Coca-Cola European Partners Great Britain Limited against the HMRC decision, in so far as that appeal relates to the Vauxhall Vivaro, is dismissed.

The Taxpayers pay HMRC's costs in the First-Tier Tribunal, the Upper Tribunal and the Court of Appeal, to be assessed if not agreed, within 28 days of a court order confirming the sums payable to HMRC being sealed. Costs are to be assessed on the standard basis. The costs orders made by the First-Tier Tribunal and the Upper Tribunal are set aside.

The Taxpayers repay HMRC the sums paid to them by HMRC pursuant to costs orders made in the First-Tier Tribunal and Upper Tribunal within 28 days of this order.