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 Herrington and Judge Brannan)
[2019] UKUT 332 (TCC)

		Royal Courts of Justice Strand, London, WC2A 2LL
		1 February 2021

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 1 February 2021 at 10.30 a.m.

Lady Justice Rose:

1. Introduction and outline of the issues

1. This appeal concerns the extent to which the Respondent, SSE Generation Ltd ('SSE'), is entitled to claim capital allowances for the expenditure that it incurred on plant when constructing the Glendoe Hydro Electric Power Scheme near Fort Augustus above Loch Ness in the Highlands of Scotland ('the Scheme'). The Scheme was opened by Her Majesty the Queen in June 2009, less than three years after the construction work first broke ground. The arguments about the tax treatment of various components that make up the project, however, still remain to be resolved over ten years after the project was completed.
2. HMRC issued closure notices to SSE for the tax years ending 31 March 2006 to 31 March 2012. They concluded that SSE's profits had been understated in those years as a result of claims made by SSE to excessive capital allowances, contrary to the provisions in Part 2 of the Capital Allowances Act 2001. SSE appealed against the notices to the First-tier Tribunal (Tax Chamber) ('the FTT'). The FTT (Judge Kevin Poole) released its decision on 31 July 2018: [2018] UKFTT 416 (TC). The FTT decided that SSE was entitled to claim allowances for some of the disputed items but upheld HMRC's view on others. HMRC lodged an appeal before the Upper Tribunal in respect of some of the items which the FTT had found did give rise to allowable expenditure. The Upper Tribunal disagreed with the FTT's analysis in certain respects and remade the decision largely in SSE's favour. The Upper Tribunal granted permission to appeal to this court on 23 December 2019.
3. Capital expenditure is not deductible from trading profits for the purpose of computing corporation tax but capital allowances can be claimed for such expenditure if it fulfils certain statutory criteria. The relevant provisions are set out in the Capital Allowances Act 2001 and references to sections, Chapters and Parts in this judgment are to that Act unless otherwise stated. Section 11 provides that allowances are available if the person carries on a qualifying activity and incurs qualifying expenditure. 'Qualifying activity' is defined in Chapter 2 of Part 2 and it is common ground that SSE is carrying on a qualifying activity.
4. Qualifying expenditure is defined by a general rule to which there are various exceptions. The general rule is set out in section 11(4). Expenditure is 'qualifying expenditure' if it is spent on plant or machinery wholly or partly incurred for the purpose of a qualifying activity. It is now also common ground that all of the items in dispute in this appeal count as 'plant' for this purpose and that they all fall within the general rule. The issue is whether they are taken out of the general rule because they fall into one of the exceptions, largely contained in Chapter 3 of Part 2. The sections in Chapter 3 that are particularly relevant for our purposes are sections 22 and 23. Section 22 contains two exceptions to the general rule. The first limb, in section 22(1)(a), excepts expenditure "on the provision of a structure or other asset in List B". Section 22(2) clarifies that the 'provision' of a structure includes its construction or acquisition. List B sets out seven groups of items and List B Item 1 includes tunnels and aqueducts. The FTT and the UT had to consider whether some elements of the Scheme are properly described as tunnels or aqueducts for this purpose. That issue is also before us. If so, the plant falls outside the general rule, meaning that no capital allowance can be claimed for the expenditure incurred in providing it. Item 7 of List B is a sweep-up item that brings within List B all structures not covered by Items 1 to 6. However, there are some structures carved out of Item 7, including structures which count as "industrial buildings". So if a structure does not fall within any of Items 1 to 6 of List B and it is an industrial building, it does not fall within List B at all and so is not excluded from the general rule by section 22(1)(a). It is common ground in the appeal that if all the items of plant in dispute are properly described as structures, then they do all fall within the definition of 'industrial buildings', so that if the plant does not constitute a 'tunnel' or an 'aqueduct' then it is not excluded from the general rule by section 22(1)(a) because it is not swept up by List B Item 7.
5. The second exception contained in section 22 is that in section 22(1)(b). That excludes "expenditure on any works involving the alteration of land". One of the important issues in the appeal is the relationship between section 22(1)(a) and (b). Are they intended to be mutually exclusive, such that if the plant in question counts as a "structure" it is either excluded or not by section 22(1)(a) without needing to consider section 22(1)(b)? Or do the two limbs overlap so that if, despite being a structure, the plant is not excluded from the general rule by section 22(1)(a), one must then go on and consider whether it falls within section 22(1)(b) and is disallowed via that route.
6. Section 22 has to be read subject to section 23. Section 23 lists expenditure which is 'unaffected' by section 22. This means that if expenditure falls within section 23 it can still be within the general rule, even if it falls within List B or section 22(1)(b). Section 23 applies to items in List C. There is a long list of items in List C ranging from advertising hoardings to zoo cages. The important ones for our purposes are List C Item 22 which is expenditure on "the alteration of land for the purpose only of installing plant or machinery" and Item 25 which is expenditure on the provision of "pipelines or underground ducts or tunnels with a primary purpose of carrying utility conduits". If any of the plant in dispute in this appeal falls within section 22(1) then it is necessary to consider whether it is saved by being a pipeline or whether some or all of the expenditure claimed is incurred in "installing" it at the site.
7. HMRC were represented before us by Timothy Brennan QC and Aparna Nathan QC and SSE was represented by Jonathan Peacock QC and Michael Ripley.

2. The items of plant in dispute

8. The different elements making up the Scheme were described by the FTT at [11] of its decision. The Scheme uses water collected over natural catchment areas covering some 75 square kilometres. The water enters the Scheme through a network of water intakes and is then channelled through about 12 km of conduits into the main reservoir. That reservoir is formed behind a concrete-faced, rock-filled dam and has a capacity of 12.7 million cubic metres. Beside the dam is an intake through which water passes from the reservoir into the headrace. The headrace is the conduit that carries the water under increasing pressure from the main intake at the reservoir to the generating equipment in the power cavern which houses the turbine and power generation equipment. The power cavern is a huge man-made void excavated from the solid rock some 600 metres below the reservoir and some 250 metres below the local ground level. It includes not only the chamber for housing the generating plant and machinery but also staff accommodation, toilets, showers and rest rooms. After the pressurised water from the headrace has served its purpose in the turbine, it runs away through the tailrace which takes the water out into Loch Ness. The tailrace also includes a short stretch of conduit that joins the outflow of the turbine to the tailrace as well as drainage and dewatering tunnels that enable the water to pass from the headrace to the tailrace without passing through the turbine, in case the turbine needs repair or maintenance.
9. The starting point for the FTT was to decide whether and how to divide up the whole Scheme into smaller items of plant in order then to apply the statute to those individual items and decide whether the expenditure incurred in respect of them was allowable or not. Judge Poole recorded at [30] that the parties were agreed that the Scheme should not be considered as a single item but looked at 'on a piecemeal basis'. He recorded at [3] and [4] of the FTT decision that he had before him a great deal of detailed material concerning the construction, repair and operation of the Scheme and that he had the benefit of a site visit which had assisted his understanding greatly.
10. Judge Poole divided the network of conduits that bring the water from the catchment areas into the main reservoir into various separate items of plant. Three of the items of plant in dispute in this appeal are conduits which are upstream of the main reservoir. The first kind is referred to as a drill and blast conduit. There are about 6 km of drilled and blasted conduit through the rock of the mountains above Loch Ness. The walls of these conduits are lined with steel mesh to reinforce them and concrete is then sprayed on to cover the steel mesh and to dry into a smoother, interior concrete wall. These conduits have an internal diameter of between 2 and 3 metres with a flat base, vertical walls up to about 1.5 metres and a semi-circular 'roof'. The second kind of conduit is described as a 'cut and cover' conduit. These are made by digging a trench in the ground, lining the trench with reinforced concrete, making semi-circular sections of concrete to form a roof over the trench so that the cross section of the conduit is roughly hexagonal with the lower half of the cross section below ground level and the upper half above ground level. Then the conduit is completely covered over with earth so that it blends into the landscape and is no longer visible from the surface of the land. There is about 1 km of 'cut and cover' conduit in the Scheme. Thirdly, there are about 800 metres of uncovered channels comprising a trench into the ground surface and lined with rocks or concrete.
11. In addition to the conduits, there is the headrace and the tailrace. The headrace is 6.2 km long and 5 metres in diameter with a 600 metre vertical drop along its length. It is entirely underground, created by boring through the rock. The FTT stated at [11(5)] that the choice of using a subterranean shaft rather than running the headrace along the surface of the ground was driven partly by engineering considerations and partly to minimise the environmental impact of the Scheme. At the foot of the headrace, the last 85 metres is substantially narrower and contains a tapering steel lining which attaches directly to the inlet valve adjacent to the turbine. The 220 metres of headrace above that part was constructed with a reinforced concrete lining inside the shaft. The main section of the headrace above that was partly stabilised with rock bolts and lined with shotcrete (concrete sprayed onto the rock surface at high pressure to strengthen the rock walls), where geological conditions require it. At the foot of the headrace, the water is under a pressure of approximately 900lb/in² and the steel-lined and reinforced concrete-lined sections prevent the water pressure from bursting through the rock and flooding the power cavern. During August 2009 it was discovered that there had been a major rockfall within the headrace about 2 km downstream from the reservoir. A bypass was constructed to take the headrace round the obstructed area and the Scheme was brought back into full operation in August 2012. It was agreed by the parties before the FTT that the expenditure on this remedial work should be treated in the same way for capital allowance purposes as the initial expenditure on the construction of the headrace: see [18] of the FTT's decision.
12. The tailrace is a little over 2 km in length. It was constructed by drill and blast for the first 340 metres of its length from Loch Ness and was bored for the remainder of its length using the same tunnel boring machine that created the headrace.

3. The legislation in more detail

13. The general rule I have described earlier is set out in section 11:

"11 General conditions as to availability of plant and machinery allowances

(1) Allowances are available under this Part if a person carries on a qualifying activity and incurs qualifying expenditure.

(2) "Qualifying activity" has the meaning given by Chapter 2.

(3) Allowances under this Part must be calculated separately for each qualifying activity which a person carries on.

(4) The general rule is that expenditure is qualifying expenditure if –

(a) it is capital expenditure on the provision of plant or machinery wholly or partly for the purposes of the qualifying activity carried on by the person incurring the expenditure, and

(b) the person incurring the expenditure owns the plant or machinery as a result of incurring it.

(5) But the general rule is affected by other provisions of this Act, and in particular by Chapter 3."

14. Section 21 in Chapter 3 of Part provides that expenditure on the provision of plant or machinery does not include expenditure on the provision of a building:

"21 Buildings

(1) For the purposes of this Act, expenditure on the provision of plant or machinery does not include expenditure on the provision of a building.

(2) The provision of a building includes its construction or acquisition.

(3) In this section, 'building' includes an asset which –

(a) is incorporated in the building,

(b) although not incorporated in the building (whether because the asset is moveable or for any other reason), is in the building and is of a kind normally incorporated in a building, or

(c) is in, or connected with the building and is in list A.

List A

Assets treated as buildings

1. Walls, floors, ceilings, doors, gates, shutters, windows and stairs.

2. Mains services, and systems, for water, electricity and gas.

3. Waste disposal systems.

4. Sewerage and drainage systems.

5. Shafts or other structures in which lifts, hoists, escalators and moving walkways are installed.

6. Fire safety systems."

15. Expenditure caught by section 21 is not inevitably prevented from being an allowable expense, taking the capital allowances regime as a whole, because there are other provisions in the Act that deal specifically with allowances for structures and buildings, for example Part 2A. The effect of section 21 is only that the expenditure is not to be treated as allowable expenditure on plant and machinery.
16. Section 22 excludes expenditure on structures, assets and works:

"22 Structures, assets and works

(1) For the purposes of this Act, expenditure on the provision of plant or machinery does not include expenditure on –

(a) the provision of a structure or other asset in list B, or

(b) any works involving the alteration of land.

List B

Excluded structures and other assets

1. A tunnel, bridge, viaduct, aqueduct, embankment or cutting.

2. A way, hard standing (such as a pavement), road, railway, tramway, a park for vehicles or containers, or an airstrip or runway.

3. An inland navigation, including a canal or basin or a navigable river.

4. A dam, reservoir or barrage, including any sluices, gates, generators and other equipment associated with the dam, reservoir or barrage.

5. A dock, harbour, wharf, pier, marina or jetty or any other structure in or at which vessels may be kept, or merchandise or passengers may be shipped or unshipped.

6. A dike, sea wall, weir or drainage ditch.

7. Any structure not within items 1 to 6 other than –

(a) a structure (but not a building) within Chapter 2 of Part 3 (meaning of 'industrial building'),

(b) a structure in use for the purposes of an undertaking for the extraction, production, processing or distribution of gas, and

(c) a structure in use for the purposes of a trade which consists in the provision of telecommunication, television or radio services.

(2) The provision of a structure or other asset includes its construction or acquisition.

(3) In this section –

(a) 'structure' means a fixed structure of any kind, other than a building (as defined by section 21(3)), and

(b) 'land' does not include buildings or other structures, but otherwise has the meaning given in Schedule 1 to the Interpretation Act 1978."

17. Section 22(2) reflects the decision of the House of Lords in Inland Revenue Commissioners v Barclay, Curle & Co Ltd [1969] 1 WLR 675. In that case it was held that the cost of excavating a new basin to create a dry dock for use in the taxpayer's shipbuilding trade was expenditure on the provision of the plant which comprised the dry dock.
18. Section 23 then limits the application of sections 21 and 22. First, section 23 (1) and (2) lists some kinds of expenditure covered by other provisions in the Act and so not to be treated as covered by sections 21 and 22. These include thermal insulation, expenditure on safety at sports grounds and software. Subsection (3) then provides:

"(3) Sections 21 and 22 also do not affect the question whether expenditure on any item described in list C is, for the purposes of this Act, expenditure on the provision of plant or machinery."

19. List C then sets out a list of 33 items, some of which contain numerous different things; for example, Item 5 covers a wide range of household goods such as cookers, washing machines, dishwashers, showers, furniture and furnishings and Item 32 covers amusement park rides. Both parties told us that the items in List C were intended to preserve the position arrived at in many past court rulings on the meaning of plant. As the Upper Tribunal commented at [66], the boundary between buildings and structures on the one hand and plant on the other hand had been eroded over a number of years by court decisions, creating uncertainty for both taxpayers and the Revenue. These provisions had been intended 'to draw a line in the sand' and entrench the understanding of the different classifications. Some items in List C were readily referable to particular cases: for example List C Item 16 covering swimming pools and diving boards reflects the decision of Megarry J in Cooke (Inspector of Taxes) v Beach Station Caravans Ltd [1973] 3 All ER 159 and Item 14 covering decorative assets provided for the enjoyment of the public in a hotel or restaurant was a response to the decision of the House of Lords in Commissioners of Inland Revenue v Scottish & Newcastle Breweries Ltd [1982] 1 WLR 322. That said, neither of the parties took us to any earlier case law dealing with items of expenditure similar to the components of the Scheme which might indicate whether a broad or narrow meaning should be given to the words in issue.
20. The two Items in List C relevant to this appeal are:

"22. The alteration of land for the purpose only of installing plant or machinery.

…

25. The provision of pipelines or underground ducts or tunnels with a primary purpose of carrying utility conduits."

21. As regards List C Item 25, the qualifying words referring to the 'primary purpose of carrying utility conduits' apply only in relation to ducts and tunnels. List C Item 25 therefore covers all pipelines but only those underground ducts and tunnels with that primary purpose: see the discussion at [74] and [75] of the FTT's decision.
22. It was agreed between the parties that all the items in dispute are 'plant' according to the common law definition of that term. Further, there was no challenge before us to the decision of the Upper Tribunal that an item which falls within List B still counts as 'plant' for the purposes of applying List C: see [133]. As I have mentioned, it was also common ground that those disputed items which count as structures but which do not fall within Items 1 to 6 of List B will not be caught by Item 7 because they are 'industrial buildings' within the meaning of Item 7(a). The term 'industrial building' is defined in section 271(1)(b) as including a structure which is in use for the purposes of a qualifying trade, and qualifying trades include the generation of electrical energy: see item 1 of Table B in section 274.

4. The relationship between section 22(1)(a) and section 22(1)(b)

23. Ground 1 of HMRC's appeal raises the issue whether the two limbs of section 22(1) are mutually exclusive in the sense that if an item of plant is a structure, then the application of section 22(1) is limited to considering whether that item is excluded from the general rule by section 22(1)(a). This means in the present case that if the application of section 22(1)(a) results in the expenditure not being excluded from the general rule, then that expenditure is allowable and one does not need to go on consider in the alternative whether it is excluded by section 22(1)(b).
24. The FTT dealt with this issue at [39] and [40]. Judge Poole preferred SSE's approach in principle, holding that section 22(1)(a) and (b) were alternatives and did not overlap: "the "works" referred to in section 22(1)(b) must be works where the alteration of land is the objective in its own right, not including works whose objective is the creation of some other asset or structure identified in List B."
25. The Upper Tribunal broadly agreed with the FTT's conclusion on this point. They correctly identified the question in [73] as whether expenditure on a 'structure', which is not a structure excluded by the operation of List B because it is an 'industrial building' saved by Item 7(a) of List B, may still not qualify for capital allowance because the expenditure comprises expenditure on works involving the alteration of land within section 22(1)(b). The Upper Tribunal noted that it was difficult to conceive of the construction of a structure which was an industrial building but which did not involve any alteration of the land. It was also, therefore, difficult to see what purpose List B served if HMRC's interpretation of the provisions was correct:

"Parliament must have known that the term "provision" when used in relation to a structure did, because of the way the term had been interpreted in case law, include any alterations of land which were necessary for the construction of the structure in question and because Parliament must have intended to ensure that the specific exceptions it provided in Item 7 in List B were effective. That intention would be thwarted were the structure in question to be excluded on the basis that its construction entailed works which involved the alteration of land."

26. The overlap would be particularly broad, if HMRC's submissions were accepted, because HMRC did not argue that expenditure could be split so that only that part of the expenditure on an industrial building arising from the alteration of the land would be covered by section 22(1)(b), leaving the remaining part as an allowable capital expense. Section 22(1)(b) would therefore cover all the expenditure on many, if not all, the structures covered by section 22(1)(a) and List B, not only those which escaped exclusion under section 22(1)(a) because they benefited from one of the three carve outs in List B Item 7. The Upper Tribunal therefore held that the two limbs of section 22(1) were mutually exclusive.

Discussion

27. In my judgment the Upper Tribunal's reasons at [79] to [82] for holding that there is no overlap between section 22(1)(a) and section 22(1)(b) are unimpeachable. Section 22(1)(b) is intended to exclude works involving the alteration of land which do not amount to structures; 'works' here being a description of the result of the alteration of the land rather than the process of altering the land. Some examples were explored during the hearing – the creation of fairways and roughs in a golf course, the creation of a bank of earth behind the targets to absorb stray bullets at a shooting range or the earth berms, hollows and runways used on a BMX track which do not include any concrete or steel structures. Mr Brennan objected that these would fall outside section 22 anyway because they would be premises rather than plant. He suggested grassland racehorse gallops as something that might be plant but not a structure, though they too might count as premises. In my view, section 22(1)(b) is there because Parliament foresaw that taxpayers might argue that the earthworks described in all these examples function as plant, given the nature of the taxpayer's trade. Section 22(1)(b) is included to forestall that because it was precisely that line between plant and premises or buildings that the provision was intended to firm up.
28. HMRC's attempt to counter the point which weighed heavily with the FTT and UT that the carve outs in List B Item 7 would be substantially eroded by any overlap with section 22(1)(b) evolved during the course of argument. Mr Brennan seemed to agree that section 22(1)(b) was intended to catch situations where the altered land itself functioned as plant. He regarded the headrace as an example of such works since it is simply a void bored through the rock. When pressed with the point that concerned the FTT and Upper Tribunal that most if not all plant intended by Parliament to benefit from the carve outs in List B Item 7 would be disallowed by section 22(1)(b), he gave two answers, using as an example the construction of a telephone mast that required substantial concrete foundations to be dug into the ground to anchor it. One answer was that the expenditure on digging the foundations might be works involving the alteration of land but would be saved by List C Item 22. The other was that those foundation works would be ancillary to the provision of the telephone mast and that would, in some way, result in them being caught only by section 22(1)(a) and not by section 22(1)(b). He contrasted this with the position of plant where the bulk of the expenditure would be on the works altering the land and only a small amount relate to the provision of some ancillary structure. What remained unclear to me was whether in putting forward these arguments he was departing from the position that the Upper Tribunal recorded at [81] that HMRC were not arguing that expenditure on items such as a mobile phone mast could be split, so that the works involving the alteration of land caught by section 22(1)(b) would be disqualified but the remaining expenditure would be saved by List B Item 7 and so allowable. In his submissions in reply, Mr Brennan appeared to accept the FTT's conclusion that section 22(1)(b) applied only where the alteration of the land was an objective in its own right and at certain points it appeared that there was no difference between him and Mr Peacock except for the debate about whether the headrace was a structure or not.
29. In my view, the difficulties that Mr Brennan encountered in trying to express HMRC's submission on the relationship between the two limbs so as to avoid the pitfall identified by the FTT and the Upper Tribunal indicate strongly that SSE's alternative interpretation is the correct one.
30. Mr Peacock also relied on the heading to section 22 "Structures, assets and works" which suggests that there are three discrete things covered by the provision, the first two covered by section 22(1)(a) and the third covered by section 22(1)(b). He referred us to Bennion on Statutory Interpretation section 16.7, citing the House of Lords in R v Montila [2004] UKHL 50, [2005] 1 All ER 113 at [34]. A heading is part of an Act and may be considered in construing any provision of the Act, provided due account is taken of the fact that its function is merely to serve as a brief guide to the material to which it relates and that it may not be entirely accurate. I agree that to that limited extent, the heading assists SSE's argument rather than HMRC's.
31. I am less convinced by the Upper Tribunal's reasoning at [83] that the mutual exclusivity of the two limbs is supported by section 22(3)(b) which provides that 'land' in the phrase 'works involving the alteration of land' in section 22(1)(b) does not include buildings or other structures. Given that the term 'land' commonly does include buildings and other structures fixed to the ground, the potential scope of section 22(1)(b) would be very broad if interpreted as covering expenditure on alterations to buildings and structures. That does not, in my view, help in deciding whether expenditure on works involving the alteration of land in the sense of the ground or earth are covered by section 22(1)(b) even if the plant they give rise to is a structure.
32. I would therefore dismiss this ground of appeal and hold that the FTT and Upper Tribunal were correct in deciding that if the plant in question is a structure or an asset then either it is excluded because it is covered by section 22(1)(a) and List B or it is an allowable capital expense (provided it is not affected by another provision of the Act). One does not need to consider whether it is caught by the exclusion in section 22(1)(b).

5. The meaning of 'tunnel' in List B Item 1

33. The FTT recorded at [35] an attempt by HMRC to introduce expert evidence on the meaning in a civil engineering context of some of the words used in the legislation. This attempt was rejected, Judge Poole holding that the words used are not specialist terms but have ordinary English meanings. There was no appeal before the Upper Tribunal on that point. Judge Poole described some of the words used in List B as being chameleons where the context can often provide colour: [38]. He therefore considered that it was legitimate to consider the way in which the items in List B had been grouped. The term 'tunnel' appeared in Item 1 alongside "bridge, viaduct, aqueduct, embankment or cutting". The grouping was significant:

"the assumption being that structures and assets which are specifically grouped together are likely to share some basic common theme and should be interpreted in accordance with that theme unless none can in fact be found."

34. Judge Poole considered the scope of the word 'tunnel' when he was analysing the expenditure on the drill and blast conduit at [66]. He referred to the Oxford English Dictionary ('OED') definitions of the words used in List B Item 1 and concluded:

"I would add that in common parlance, the word "tunnel" would normally refer to a passage bored through ground which permits people or forms of transport to pass to and fro. "Embankment" is defined in the OED as "a mound, bank, or other structure for confining a river, etc. within fixed limits" or, more familiarly, as "a long earthen bank or mound, esp. one raised for the purpose of carrying a road or a railway across a valley.". "Cutting" is relevantly defined as "an open, trench-like excavation through a piece of ground that rises above the level of a canal, railway, or road which has to be taken across it". On this basis, there does seem to be a clear theme emerging in Item 1 of structures related to transportation infrastructure."

35. He concluded that the drill and blast conduit did not comprise a tunnel within the ordinary meaning of the word because although it was large enough to allow a person to enter it and pass from one end to the other, that was not any part of its intended purpose.
36. The Upper Tribunal arrived at the same narrower interpretation of the word 'tunnel' but by a different route. They held that the FTT had made an error of law in concluding that the ordinary meaning of the word was limited to subterranean passageways which are intended to be used for the transportation of people or forms of transport. This was contrary, the Upper Tribunal held, to the dictionary definition that makes it clear that the term can embrace any form of subterranean passageway. However, they then applied the principle of construction known by the Latin maxim res noscitur a sociis, that is, that the meaning of a word is affected by the words immediately surrounding it: [93]

" … in our view the words immediately surrounding "tunnel" in Item 1 of List B are "bridge, viaduct, aqueduct, embankment or cutting" all of which are the product of civil engineering works related to the construction of transportation ways and routes, that is the types of ways and routes which the draftsman subsequently lists in Item 2 and 3 of List B. It follows therefore that the context requires that the word "tunnel" should be given a narrower meaning than its ordinary dictionary meaning."

37. They therefore concluded that in its context, the word 'tunnel' had a narrower meaning than its ordinary meaning.

Discussion

38. The most relevant OED definition of the word 'tunnel' is "a subterranean passage; a road-way excavated under ground, esp. under a hill or mountain, or beneath the bed of a river: now most commonly on a railway; also in earliest use on a canal, in a mine etc"
39. On this point I have concluded that the FTT was right to hold that the ordinary meaning of the word 'tunnel' in this context denotes something along which people or vehicles are intended to travel and not simply any subterranean passage. That limits the term to a subterranean conduit of a diameter and gradient which enables it to operate as a 'passageway' as indicated by even the broadest OED definition. The reference to gradient is prompted by a point made by Popplewell LJ during the hearing that one would not normally describe a vertical shaft drilled through rock, for example a mine shaft, as a tunnel. I agree with Judge Poole's comment at [38] that some of the words used in List B have an elastic meaning and that they can take on the colour of the words surrounding them. Determining the scope of such a chameleon word is not simply a binary choice between the widest possible dictionary meaning on the one hand and a narrowed meaning, specific to a particular statutory provision derived from the noscitur a sociis interpretative tool on the other - it is a more nuanced exercise than that. Many English words have a number of ordinary meanings in common usage and the statutory meaning is not necessarily the broadest one.
40. We were referred to three cases in which the courts have applied the interpretative tool that res noscitur a sociis. In Pengelley v Bell Punch Co Ltd [1964] 2 ALL ER 945, an employee was injured when reaching past large reels of paper stored on the floor of a factory to get to smaller reels stored on racks above. He alleged that by placing the large reels on the floor, his employer was in breach of an obligation imposed by the Factories Act 1961 to keep floors, steps, stairs, passages and gangways free from any obstruction. Diplock LJ observed that the last four locations were places used for the purpose of passage. The reference to 'floors' in that context and in the light of the word 'obstruction', meant that the obligation was limited to keep clear those parts of the factory floor on which workmen were likely to pass and repass. Where a part of the factory floor was properly used for storing as it was in this case, the articles stored were not obstructions on the floor for this purpose. In Shamoon v Chief Constable of the Royal Ulster Constabulary [2003] UKHL 11, [2003] 2 ALL ER 26, the issue was the meaning of the word 'detriment' in an anti-discrimination provision which made it unlawful for a person to discriminate against a female employee (a) in the way he afforded her access to opportunities for promotion, transfer or training, or to any other benefits, facilities or services, or by refusing or deliberately omitting to afford her access to them, or (b) by dismissing her, "or subjecting her to any other detriment". The placing of the word 'detriment' showed that the disadvantage complained of by the employee had to arise in the employment field: see [34] per Lord Hope of Craighead. Finally, in Tektrol Ltd (formerly Atto Power Controls Ltd) v International Insurance Co of Hanover Ltd and another [2005] EWCA Civ 845, [2006] 1 All ER 780, a clause in an insurance policy excluded the insurer's liability for damage caused by erasure, loss, distortion or corruption of information on computer systems caused deliberately by rioters, strikers, locked-out workers, persons taking part in labour disturbances or civil commotion or malicious persons. The insured's software code had been lost partly because of a virus sent by an unknown person under the guise of an emailed Christmas card and partly because burglars had broken into the premises and stolen the computers. Buxton LJ held at [11] and [12] that although the author of the emailed virus was certainly a "malicious person" in the ordinary sense, the exclusion was limited by the previous references to rioters and locked-out workers etc to people who destroyed the computer systems on or near the insured's premises: "suddenly to tag on at the end of the excepting clause a reference to remote hackers, a completely different category of person making a completely different kind of attack, significantly changes the thrust of the exception, in a way that one would expect to be done only by much more specific wording". Carnwath LJ agreed with Buxton LJ on this point, as did Sir Martin Nourse. Sir Martin referred expressly to the noscitur a sociis principle when construing the word 'loss' in the phrase "other erasure loss distortion or corruption of information". He held at [29] that the word 'loss' there meant only loss by means of electronic interference and not loss by theft of the computers.
41. Those cases employed the noscitur a sociis tool of construction to do something more than choose between alternative ordinary meanings of a particular word. No one could suggest that the large reels of paper in Pengelly had been placed on a surface that was not within any ordinary meaning of the word 'floor' and no one would suggest that a possible ordinary meaning of the phrase 'malicious person' would be limited to persons who were in or near a particular location. A detriment suffered in the employment field is an example of a detriment but that is different from it being one of a number of possible ordinary meanings of the word 'detriment'. One must not confuse the fact that there is more than one dictionary definition of a word with the fact that there are many different kinds or examples of the thing itself. A word in its statutory context may by the application of the noscitur a sociis principle be interpreted as limited to a specific example of the thing. The principle was used in the cases cited to cut down the scope of the word to give it a much narrower meaning than its ordinary meaning – it had a meaning bespoke to the context in which it was used.
42. I agree with Judge Poole and with the Upper Tribunal that it is significant that the structures or assets in List B are grouped together into particular themes and that the theme emerging from Item 1 is a transportation theme. Item 2 also has a transportation theme; as Mr Peacock put it, it includes those things, the road, railway and tramway, which go through or over the things in Item 1. Items 3, 4 and 5 are all water related in some way. The ordinary meaning of the word 'tunnel' in this context is, as the FTT held at [66], "a passage bored through ground which permits people or forms of transport to pass to and fro". One does not, in my judgment, have to rely on the noscitur a sociis principle to arrive at that conclusion. That does not rule out the possibility that the word 'tunnel' in a different statutory context might bear the wider ordinary meaning or might to the contrary be cut down to a particular example or kind of a tunnel.
43. HMRC argued that the word 'tunnel' in List B could not bear the limited meaning given to it by the FTT or Upper Tribunal because the word appears not only in List B Item 1 but in List C Item 25. Item 25 refers to tunnels with the primary purpose of carrying utility conduits. That use of the word 'tunnels' was clearly not limited to tunnels for transportation but the draftsman, Mr Brennan submitted, would not have used the same word with different meanings in such close proximity. The FTT rejected this argument at [67], holding that the tunnel referred to in List C Item 25 would have a transportation purpose but that would not be its primary purpose. The Upper Tribunal rejected the same argument on the basis that it was clear that the term was being used in a different sense in List C Item 25, namely to denote a subterranean passageway which does not have a transport function: [96].
44. I do not consider that the use of the word 'tunnel' in List C Item 25 necessarily bears the same meaning as it bears in List B Item 1. I do not need to decide the point at which an underground duct becomes a tunnel since both the ducts and the tunnels must have the primary purpose of carrying utility cables before they benefit from the saving. If the underground conduit is wide enough to accommodate people and vehicles and is intended to do so, it may be better described as a tunnel than a duct. In either event it will be covered by List C Item 25 only if its primary purpose is the purpose specified even though its secondary purpose may be to provide passage for people and vehicles.
45. Mr Brennan criticised the Upper Tribunal for failing to grapple with the question why Parliament would have wanted as a matter of policy to draft List B Item 1 so as to exclude only transport tunnels and not other subterranean passages. I would counter his rhetorical question with another – why would Parliament take the trouble to exclude from List B those items saved by item 7 (b) for the benefit of undertakings engaged in gas extraction, production, processing or distribution of gas and yet not save structures comprising subterranean passages in use by that undertaking? Mr Peacock pointed out in argument that the kinds of structures which come within Item 7(a) as industrial buildings include structures used for the purpose of an undertaking for the supply of water for public consumption or for the provision of sewage services. When trying to discern the dividing lines drawn by detailed taxing provisions one cannot rely on broad policy goals as an aid to interpreting where the line should be drawn: see HMRC v Trigg [2016] UKUT 165 (TCC), [2016] STC 1310, [35].
46. I would therefore dismiss Ground 2 of HMRC's appeal.

6. The meaning of 'aqueduct' in List B Item 1

47. The FTT considered the meaning of 'aqueduct' when applying the provisions to the drilled and blasted conduits. Judge Poole noted that there were two OED definitions of the word. It can mean "an artificial channel for the conveyance of water from place to place; a conduit; esp an elevated structure of masonry used for this purpose" or it can mean "The similar structure by which a canal is carried over a river etc." He held ([69]):

"the better view to be that in the context of List B, the word "aqueduct" is apt to describe an asset of the type we are here concerned with – an artificial underground conduit whose function is solely to transport water from one place to another through the ground under the force of gravity; I consider the transportation of water itself is enough to be consistent with the overall "transportation" theme of Item 1, rather than requiring the water to be the means of transportation of other things (as in the case of a canal)."

48. The Upper Tribunal considered the meaning of 'aqueduct' at [97] onwards. They agreed with the FTT that the word had two potential ordinary meanings according to the OED, one broader and one narrower. However, the Upper Tribunal held that the transportation theme of List B Item 1 meant that the second meaning was the one intended here. The FTT had erred in holding that the fact that some of the items in dispute in the appeal moved water from place to place was enough to make them aqueducts. The Upper Tribunal confined the term aqueduct to a bridge-like structure which created a transportation route in the form of a canal: [101].

Discussion

49. I have found this Ground more difficult since the kind of structure that comes immediately to mind as being an aqueduct is neither an underground nor surface channel for moving water, nor a bridge like structure carrying a canal for navigation. It is a bridge-like structure which spans a river, gorge or valley in order to carry either a general water supply from a distant source to a centre of population or a canal. The definition used by the Upper Tribunal would exclude many spectacular and elegant structures that have been left to us from ancient times and are commonly thought of and referred to as aqueducts. It would include only those structures of which there are a number across the country which carry canals. However, I am satisfied that that is right and that the transportation theme of Item B List 1 limits the scope of the word here to canal-carrying aqueducts. Interesting though it might be to wonder whether the builders of the Pont du Gard in France or the Gadara Aqueduct in modern day Syria would have been able to claim capital allowances for the costs incurred, had the Roman Empire's legal system included a provision similar to section 22(1)(a), that is not likely to be a question to which Parliament was addressing its mind in putting together List B Item 1. This is a provision where the noscitur a sociis principle of interpretation comes into play to limit the ordinary meaning of the word to a particular kind of aqueduct, namely that with a transportation theme. I therefore agree with the Upper Tribunal that in its context here, an aqueduct is a bridge or viaduct-like structure which carries a canal.
50. HMRC point out that if the term has such a limited meaning, there was no reason to include it in List B Item 1 because it is already included in Item 3, which refers expressly to canals as an example of inland navigation items. I do not consider that undermines the Upper Tribunal's reasoning. There does not appear to have been any attempt by the drafter of the Lists in sections 21, 22 and 23 to avoid duplication and I do not accept that one must treat the meaning of a word in one item of the List as being narrowed by the inclusion of a similar word in another.
51. In any event, the inclusion of aqueducts in Item 1 may have been intended to forestall any argument that Item 3 related only to the canal itself and not to the bridge-like structure which carries it and which properly belongs in Item 1. Similarly, a viaduct is a particular kind of bridge but is included, in addition to bridges, in Item 1 even though the railway or road which is carried by the viaduct is then separately included in List B Item 2.

7. The application of the provisions to the disputed plant

52. The FTT held that drill and blast conduits, the 'cut and cover' conduits and the open channels were not tunnels because they were not intended to be used for the passage of people or transport: see [67], [82] and [86]. They were all aqueducts and so caught by List B Item 1: see [69], [82] and [86]. Similarly, Judge Poole held that the tailrace was an aqueduct though not a tunnel: [113].
53. Judge Poole therefore went on to consider whether these items of plant fell within List C Items 22 or 25 so that they might still attract capital allowances despite falling within section 22(1)(a). He held that Item 25 did not apply to any of the conduits since none of the conduits was a pipeline. His conclusions as to the application of List C Item 22 differed as between the different kinds of conduits. As regards the drill and blast conduits, he regarded the sole purpose of the drill and blast process as being, in essence, to alter the land in order to form the conduit. Whether that meant that the expenditure on drilling and blasting was expenditure for the purpose only of installing the conduit depended, therefore, on whether the term 'installing' included creating the item of plant in situ as well as its more common meaning of putting in place an item of plant that existed before coming onto the site, albeit in component form: [78]. He concluded that 'installing' did extend to creating the plant in situ and that all of the expenditure on creating the drill and blast conduit fell within List C Item 22. That expenditure was therefore allowable in full. He came to a similar conclusion in respect of the open channel conduits since the process of creating the aqueduct by excavating rough channels and lining them with rocks or concrete was all expenditure incurred in altering land for the sole purpose of installing the resulting aqueducts: [87]. All the expenditure was therefore allowable.
54. The same reasoning applied to the expenditure on the tailrace which was therefore allowable in full: [113].
55. As regards the 'cut and cover' conduits, Judge Poole noted the difference in construction between these and the drill and blast conduits. The erection of the concrete aqueduct structure in these conduits was not an alteration of land so the costs of that erection were not saved by Item 22. However, he held that the costs of excavating its base and subsequently covering it over were costs incurred in the alteration of land for the purpose only of installing the aqueduct structure. He concluded:

"85. It follows that no allowances are available for the expenditure on the fabrication in situ of the concrete conduit itself (excluded as an "aqueduct" by Item 1 in List B), but the expenditure incurred on the preparatory excavations and the subsequent re-covering of the conduit after it had been built is allowable. Whilst it might appear a somewhat counter-intuitive result that the method of construction should make a difference to the CAA treatment in this way, intuition is rarely a reliable guide to statutory interpretation, and the difference in treatment flows logically, in my view, from the terms of the legislation."

56. In relation to the drill and blast conduits he held that section 22(1)(b) did not apply but he said that if he had not found that it was an aqueduct he would instead have found it to be works involving the alteration of land within section 22(1)(b): [70].
57. As regards the headrace, Judge Poole held ([91] – [101]) that the functions of the headrace were very different from those served by the conduits that he had concluded were aqueducts. He did not consider that the headrace was an aqueduct because its function was far more complex than simply transporting water from one place to another. It was designed to deliver the volume and pressure of water needed for the turbine without allowing it to escape or burst through to the power cavern. The headrace was also not a tunnel, applying the limited meaning Judge Poole had given to that term in List B Item 1. Having found that it was not within List B, he went on to consider whether it was covered by section 22(1)(b). He concluded that some of the expenditure on the headrace was incurred on "works involving the alteration of land": [94]. He held that the costs of drilling the headrace, reinforcing and lining it were expenditure caught by section 22(1)(b) and would be disallowed to that extent unless saved by List C. HMRC had accepted that the cost of the tapered steel pipe and of boring the length of the headrace which contained the steel pipe did qualify for allowances. Judge Poole held that the remainder of the expenditure also qualified for allowances as having been incurred for the purpose only of installing the headrace. It benefited from List C Item 22: [100]. He held that the provisions applied in the same way to the turbine outflow tunnel and the drainage and dewatering tunnels which were adjuncts to the tailrace.
58. The Upper Tribunal agreed with the FTT that none of the plant fell within the term 'tunnel' as properly construed. They differed from the FTT in concluding that the conduits and the tailrace were not aqueducts either. Since the Upper Tribunal concluded that the conduits, the headrace and the tailrace were all structures, it followed that they were not caught by section 22(1)(b) since that limb of section 22 only applied to plant that was not a structure or asset. There was no need therefore to consider whether the conduits were 'pipelines' for the purpose of List C Item 25 though they agreed with the FTT that they were not.

Discussion

59. I agree with the Upper Tribunal's conclusion that the disputed items are neither tunnels nor aqueducts. Further, they are not caught by List B Item 7 because they are industrial buildings. Given that they are structures, they are not covered by section 22(1)(b). The expenditure on them is therefore allowable.
60. In Ground 6 of its grounds of appeal, HMRC argues that the Upper Tribunal was not entitled to differ from the FTT by reversing the FTT's decision that the headrace, tailrace and 'cut and cover' conduits all resulted from works involving the alteration of land and were excluded from allowances by section 22(1)(b). This, HMRC argues, was a question of fact. Similarly in Ground 4 they argued that in so far as Judge Poole indicated that if he was wrong about the conduits being aqueducts, then he would have held that they were caught by section 22(1)(b), that was also a finding of fact with which the Upper Tribunal could not interfere.
61. I do not accept those submissions. The FTT's application of section 22(1)(b) was inconsistent with its own earlier analysis of where the boundary between the two limbs of section 22 lies. The FTT held at [40] that the works referred to in section 22(1)(b) are only those where the alteration of the land does not result in a structure or asset identified in List B. The Upper Tribunal agreed and I have concluded that that is the right analysis. It was therefore a mistake for the FTT to apply section 22(1)(b) to the headrace or tailrace on the basis that they were not tunnels or aqueducts and were not caught by List B Item 7. The items were still structures and so not within section 22(1)(b).
62. Ground 2 of HMRC's appeal contends that the Upper Tribunal was not entitled to interfere with the FTT's findings of fact as to whether the different kinds of conduits were aqueducts. Mr Brennan relied on Clark v Perks [2001] STC 1254 to argue that the question whether the conduit was an aqueduct or not was a question of fact and not of law and could only be overturned on Edwards v Bairstow grounds (see Edwards (Inspector of Taxes) v Bairstow [1956] AC 14 at 36). SSE had not mounted any such challenge to the finding.
63. The Upper Tribunal referred to Clark v Perks at [54] onwards citing the passage from the speech of Lord Simon of Glaisdale in Ransom v Higgs [1974] STC 539 at 561:

"The meaning of a word or phrase in an Act of Parliament is a question of law not fact; even though the law may then declare that the word or phrase has no statutory meaning beyond its common acceptance and that it is a question of fact whether the circumstances fall within such meaning (Cozens v Brutus). But many words and phrases in English have many shades of meaning and are capable of embracing a great diversity of circumstance. So the interpretation of the language of an Act of Parliament often involves declaring that certain conduct must as a matter of law fall within the statutory language (as was the actual decision in Edwards v Bairstow); that other conduct must as a matter of law fall outside the statutory language; but that whether yet a third category of conduct falls within the statutory language or outside it depends on the evaluation of such conduct by the tribunal of fact. This last question is often appropriately described as one of 'fact and degree'."

64. In Clark v Perks the question was whether a mobile offshore drilling unit was a 'ship'. The General Commissioners had held that the rigs fell within the relevant definition of a ship. On the Revenue's appeal, Ferris J held that the meaning of the word 'ship' was a matter of law on which he was entitled to give effect to his own view. He held that the rigs were not ships. On further appeal the question arose whether the judge could only have interfered with the Commissioners' conclusion if it had been one that could not reasonably have been reached on the basis of the primary facts as found. Carnwath J (with whom Longmore and Robert Walker LJJ agreed) held that once the meaning of the term 'ship' had been established, the question whether the facts found brought the case within that meaning was not a question of law but a question of fact subject only to the conclusion falling within the limits of reasonableness: [29]. However, he then went on to consider whether the Commissioners' conclusion as to the meaning of the word 'ship' had been wrong in law. He held that the Revenue had not shown that the Commissioners had erred in law and allowed the taxpayers' appeal.
65. I do not see that Clark v Perks assists HMRC in this appeal. The FTT's conclusion that some of the items were conduits was the result of an error of law in their construction of that word in List B Item 1. Mr Brennan accepted that if as a matter of law, aqueducts must look like bridges in this context, he could not maintain that finding of fact as regards the conduits. That is the case here.
66. Mr Brennan also argued that the headrace at least was not a "structure" and so could be caught by section 22(1)(b) even if the two limbs of the section were mutually exclusive as SSE contended. That submission cannot be maintained, given Judge Poole's description of the headrace. It is not right to describe the headrace as simply a void bored through the rock, it is a much more complex item than that in its construction with the steel mesh, concrete and bolt reinforcement. The Upper Tribunal put the matter beyond doubt at [145]:

"145. Although the FTT did not say so in terms, in our view it is implicit in the Decision that the FTT proceeded on the basis that it had found that all the assets in dispute were "structures": see [43] of the Decision where the FTT referred to it being common ground that none of the "structures" involved in this case fall into List B by virtue of Item 7 because of the exception for industrial buildings referred to below. Alternatively, in so far as it is necessary to make an express finding of fact to that effect, we do so by the exercise of our powers under s 12(4)(b) TCEA on the basis of the evidence that we were shown as to the features of the various assets and their method of construction."

67. The FTT's conclusion that the headrace and tailrace were covered by section 22(1)(b) was an error of law, not a finding of fact, and it is one which the Upper Tribunal was entitled to correct (subject to Ground 8 of HMRC's appeal in respect of the 'cut and cover' conduits).
68. In the light of my conclusions on the meaning of tunnel and aqueduct and as to the relationship between section 22(1)(a) and (b), I do not need to address the scope of List C Item 22 and whether the creation of a plant on site amounts to 'installing' that plant so as to make it allowable as expenditure on work carried out solely for the purpose of installing the plant. The Upper Tribunal similarly did not need to decide whether any of the expenditure in dispute was incurred in altering land solely for the purpose of installing the plant. However, the Upper Tribunal reviewed the case law in detail and drew from it the common theme that installation is a "process which involves the integration, often with a degree of complexity of an article or articles which have already been made into another article, structure, building or even the land itself. In none of the cases that they had been referred to had the term been held to include the creation of an item of plant in situ": [127]. List C Item 22 was therefore confined to expenditure on items of plant which by their nature are constructed separately and then need to be installed. They concluded at [126] that the FTT had erred in holding that where an item of plant is created on site, the expenditure incurred in altering land as part of its creation could qualify for allowance on the basis that it was incurred solely for the purpose of installing that plant.
69. Although I see the force of the Upper Tribunal's reasoning and do not dissent from it, it is undesirable given the elastic nature of the words used in these provisions to come to a concluded view of their scope effectively in the abstract. List C Item 22 is intended to be applied in relation to expenditure on plant which would otherwise be caught by section 21 or 22. When one is considering an item of plant that does fall within section 21 or 22, it may be easy or difficult to determine whether there has been an alteration of the land for the purpose only of installing that item. I doubt that it is helpful to that future exercise to consider how the words would apply in relation to plant to which the words were not intended to apply.

8. The effect of SSE's Respondent's notice

70. The issue raised in Ground 8 of HMRC's appeal arises from the Upper Tribunal's disagreement with the FTT's analysis of the application of the provisions to the 'cut and cover' conduits. The FTT held that the conduit was an aqueduct and that only part of the expenditure – that incurred in excavating the trench for the conduit but not that incurred in fabricating the conduit on site – was allowable as expenditure incurred solely for the purpose of installing the aqueduct within the meaning of List C Item 22. The Upper Tribunal held that the definition of aqueduct in List B Item 1 was narrower and that the conduit was not an aqueduct. The expenditure was therefore allowable in full, without the need to rely on section 23 and List C. The Upper Tribunal noted at [45] that HMRC objected to SSE raising this point because SSE had not applied for permission to appeal against the FTT's ruling. The Upper Tribunal rejected this submission and held at [46] that SSE did not in the circumstances of this appeal need permission. The Upper Tribunal said:

"In circumstances where a respondent is opposing an appeal but wishes to challenge a finding of the FTT on a point which it argued before the FTT but in respect of which it was unsuccessful, it can do so simply through the medium of the Respondent's Notice which it may (but need not) file pursuant to Rule 24 of The Tribunal Procedure (Upper Tribunal) Procedure Rules 2008. If a Respondent's Notice is filed, then Rule 24 (3) (f) requires the Respondent to set out the grounds on which the Respondent relies in the Upper Tribunal proceedings, which the rule states are to include "any grounds on which the respondent was unsuccessful in the proceedings which are the subject of the appeal, but intends to rely in the appeal". SSEG clearly met the requirements of that rule in its Respondent's Notice, specifying a number of arguments which it ran before the FTT and which it wished to argue again. In our view, there is nothing in the rule that indicates that the rule is limited to arguments which simply seek to uphold the FTT's decision on different grounds. Therefore, provided the argument was one that was before the FTT it is open to a respondent to argue the point again in the Upper Tribunal provided it gives notice of that in its respondent's notice without needing to apply for permission to appeal."

71. The Upper Tribunal went on to say that there was no prejudice to HMRC since the argument had been squarely before the FTT and HMRC had been able to make submissions on it during the appeal.
72. On this point I respectfully differ from the analysis of the Upper Tribunal. I would uphold their conclusion that the 'cut and cover' conduit expenditure is in principle allowable in full, but would hold that it was not open to SSE to assert that, without having sought and obtained permission to appeal against the exclusion of the expenditure on fabricating the conduit.
73. The right to appeal from the FTT to the Upper Tribunal is provided by section 11 of the Tribunals, Courts and Enforcement Act 2007 ('TCEA'). That provides so far as relevant:

"11 Right to appeal to Upper Tribunal

(1) For the purposes of subsection (2), the reference to a right of appeal is to a right to appeal to the Upper Tribunal on any point of law arising from a decision made by the First-tier Tribunal other than an excluded decision.

(2) Any party to a case has a right of appeal, subject to subsection (8).

(3) That right may be exercised only with permission (…)

(4) Permission … may be given by

(a) the First-tier Tribunal, or

(b) the Upper Tribunal,

on the application by the party."

74. Section 12 TCEA then provides that if, in deciding an appeal under section 11, the Upper Tribunal finds that the making of the decision concerned involved an error of law, it may (but need not) set aside the decision and remake it. In exercising that power, the Upper Tribunal may make any decision that the First-tier Tribunal could make if the First-tier Tribunal were remaking the decision and may make such findings of fact as it considers appropriate.
75. The procedure to be followed by a party seeking permission to appeal is set out in the Tribunal Procedure (First-tier Tribunal) (Tax Chamber) Rules 2009 (SI 2009/273) ('the FTT Rules'). Rule 39 provides that a person seeking permission to appeal must make a written application to the Tribunal for permission within 56 days of the date on which the tribunal sends the relevant decision notice. If permission is refused, the party may renew the application to the Upper Tribunal pursuant to rule 21 of the Tribunal Procedure (Upper Tribunal) Rules 2008 (SI 2008/2698) ('the UT Rules'). Rule 21 provides that a person may apply to the Upper Tribunal for permission to appeal against the decision of the First-tier Tribunal only if they have applied to the First-tier Tribunal and the application has been refused or not admitted or granted only on limited grounds. According to rule 21(4) of the UT Rules, the application for permission must state amongst other things the details of the decision challenged and the grounds on which the appellant relies. If the Upper Tribunal refuses permission, it must send written notice of the refusal to the appellant: rule 22(1) of the UT Rules. If it gives permission to appeal then it sends written notice of the permission to each party, the application for permission to appeal stands as the notice of appeal and the Upper Tribunal must send each respondent a copy of the application for permission and any documents provided with it by the appellant: rule 22(2)(a) and (b) of the UT Rules. Where the FTT has already given permission to appeal to the Upper Tribunal, the appellant must provide a notice of appeal to the Upper Tribunal and the Upper Tribunal then sends a copy of that notice and any accompanying documents to each respondent: rule 23 of the UT Rules.
76. Rule 24 of the UT Rules then provides for the response to the notice of appeal. Any response must state, amongst other things:

"24(3) (f) the grounds on which the respondent relies, including (in the case of an appeal against the decision of another tribunal) any grounds on which the respondent was unsuccessful in the proceedings which are the subject of the appeal, but intends to rely in the appeal;"

77. SSE did not seek permission to appeal from the FTT and Mr Peacock accepts that they could not have sought permission from the Upper Tribunal without first having made an unsuccessful application to the FTT. I agree with HMRC's submission that rule 24(3)(f) cannot obviate the need for permission set out in section 11 TCEA. The grounds referred to there are the grounds on which the party relies in its character as a respondent to appeal. Certainly if the respondent succeeded on an issue before the FTT because the FTT accepted one of a number of arguments while rejecting other arguments for the same result, the respondent can raise those unsuccessful arguments if its success is challenged on appeal by the opposing party. But the respondent cannot raise an issue which it lost before the FTT unless it obtains permission to appeal for itself.
78. Mr Peacock argued that the value of the expenditure disallowed by the FTT in respect of the fabrication on site of the 'cut and cover' conduit was comparatively small. It would not have been worth SSE's while to appeal that point in isolation. However, once the appeal was mounted by HMRC, SSE wished to argue that the FTT erred in disallowing that part of the expenditure. It would be wrong, he submitted, to require a respondent in SSE's position to have to lodge its own appeal, without knowing whether the other party is going to bring an appeal first. There is no requirement in the FTT Rules that the opposing party is notified of an application for permission having been made by a party. By the time the respondent is made aware that a successful application has been made either to the FTT or to the Upper Tribunal, Mr Peacock pointed out, the time limit in rule 39 of the FTT Rules may well have passed. That would deprive the respondent of the opportunity then to lodge its own appeal. Mr Peacock submitted that HMRC's concern that a broad interpretation of rule 24(3)(f) of the UT Rules would entitle a respondent to re-run every issue on which it lost before the FTT without the filter of the permission stage was alleviated by the Upper Tribunal's case management powers conferred by rule 5 of the UT Rules. The Upper Tribunal must exercise those powers in a way which gives effect to the overriding objective in rule 2 of dealing with cases fairly and justly, including by dealing with the case in ways which are proportionate. In the present case, the Upper Tribunal expressly considered whether HMRC had been prejudiced by SSE being able to argue that all the expenditure incurred in the 'cut and cover' conduit and concluded that they had not. Mr Peacock referred to the 'venerable principle' that the task of the FTT and the Upper Tribunal is to arrive at the collection of the correct amount of tax: see Investec Asset Finance plc and another v HMRC [2020] EWCA Civ 579 [60] and [100]. He argued that the Upper Tribunal was therefore entitled to give effect to the logic of its conclusion that the FTT had erred in finding that the conduit was an aqueduct.
79. Although I accept that a party who has lost on a minor issue may well find itself in the same position as SSE, neither that argument based on practicalities nor the venerable principle can override the statutory requirement for permission to appeal. The procedure established by section 11 TCEA, the FTT Rules and the UT Rules is different from the procedure which operates under the Civil Procedure Rules as set out in CPR 52.13. In that rule, a respondent may serve a respondent's notice which seeks permission to appeal from the appeal court as well as asking the appeal court to uphold the decision of the lower court for reasons different from, or additional to, those given by the lower court. The respondent does not therefore have to seek permission first from the lower court within the time limit set for an initial appeal. According to the different procedure adopted under the tribunal rules, the respondent cannot seek permission to appeal in the response notice served under rule 24 of the UT Rules. A respondent in the position of SSE which, once an appeal is on foot, wants to reverse a point decided against it in the FTT must apply for permission to the FTT. If the time limit for doing so has expired, it must request an extension of time, giving the reasons why the application notice was not provided in time: see FTT Rule 39(4). At that stage the FTT will consider whether the proposed appeal meets the test for the grant of permission and whether time should be extended. The latter point will require consideration of how far the respondent's appeal will enlarge the scope of the appeal and whether it is consistent with the overriding objective to grant permission. The fact that the respondent's application would open up several new fronts in the appeal leading to a longer and more complicated hearing, does not rule out the grant of permission. The original appellant is not entitled to insist that the scope of the appeal remains within the limited compass of the grounds that it has raised. The fact that a late application for permission may widen the scope of the appeal is a risk that the appellant takes, if the respondent has properly arguable issues that could result in the appellant being worse off than if they had let the FTT's decision lie. A similar issue was considered by the Upper Tribunal (Nugee J and Judge Nowlan) in Price and others v HMRC [2015] UKUT 164 (TCC), [2015] STC 1975. I respectfully agree entirely with the analysis and reasoning set out there.
80. In considering whether a point raised in a respondent's notice can only be made if permission to appeal is granted, one must identify what decision of the FTT is being challenged. The outcome in relation to the 'cut and cover' conduits was the consequence of the Upper Tribunal having identified an error of law in the FTT's interpretation of the word 'aqueduct' in List B Item 1. That issue was before it because the grounds of appeal raised by HMRC in its appeal from the FTT to the Upper Tribunal challenged the FTT's decision that the headrace was not an aqueduct. In its response to the appeal filed with the Upper Tribunal, SSE submitted that the FTT should have concluded that the conduits and tailraces were not aqueducts. The conclusion that the drill and blast conduits, the uncovered channel conduits and the headrace are not aqueducts leads to the same result as the FTT arrived at for other reasons - the expenditure on them is allowable in full. Applying the narrower definition of 'aqueduct' to the 'cut and cover' conduits leads to a different result because the FTT allowed only part of the costs. The decision challenged here is not as to the meaning of the word 'aqueduct' but as to whether HMRC's closure notice was correct in disallowing the capital expenditure incurred on the 'cut and cover' conduits. The Upper Tribunal's decision increased the amount of allowable expenditure but that result could only be achieved if SSE had sought permission to do better than the partial allowance. No such permission had either been sought or granted and in my judgment HMRC are right to say that the Upper Tribunal erred in concluding at [161] that the expenditure was recoverable in full.

9. Conclusion

81. In the light of my conclusions on the grounds of appeal, I would dismiss HMRC's appeal save that it is allowed in relation to the Upper Tribunal's conclusion that all the expenditure on the 'cut and cover' conduits is allowable.

Lord Justice Popplewell:

82. I agree.

Lord Justice David Richards:

83. I also agree.