BAILII [Home] [Databases] [World Law] [Multidatabase Search] [Help] [Feedback]

First-tier Tribunal (Tax)

You are here: BAILII >> Databases >> First-tier Tribunal (Tax) >> Cheshire Cavity Storage 1 Ltd and EDF Energy (Gas Storage Hole House) Ltd v Revenue & Customs (PROCEDURE : Other) [2019] UKFTT 498 (TC) (01 August 2019)
Cite as: [2019] UKFTT 498 (TC)

[New search] [Context] [View without highlighting] [Printable PDF version] [Help]




[2019] UKFTT 498 (TC)

CAPITAL ALLOWANCES - gas cavities – whether plant – no – whether excluded from allowances under s 22 CAA -yes







Appeal number:  TC/2017/01148 and TC/2017/01153

























Sitting in public at Taylor House, Rosebery Avenue, London on 19-22 November 2019


Mr J Peacock, QC, instructed by Enyo Law LLP, for the Appellant


Ms A Nathan, counsel, instructed by the General Counsel and Solicitor to HM Revenue and Customs, for the Respondents




1.             This appeal concerns the tax treatment of expenses incurred in creating gas cavities.  Cheshire Cavity Storage 1 Limited was the first appellant and I will refer to it as CCS. The second appellant was EDF Energy (Gas Storage Hole House) Limited which I shall refer to as EDF.  I will refer to the two appellants collectively as ‘the appellants’. 

2.             CCS appealed against conclusions in 3 closure notices dated 3 August 2016 which amended its tax returns for the periods ended 31 December 2011 to 31 December 2013.   EDF appealed against closure notices also issued on 3 August 2016 which amended its tax returns for the accounting periods ended 31 December 2006 to 31 December 2013.

3.             The effect of the closure notices was to deny both companies capital allowances in respect of the cost of de-brining gas cavities, and to deny EDF capital allowances in respect of the cost of leaching gas cavities.

4.             I am asked only to make a decision in principle on whether the appellants were entitled to capital allowances on these costs; the parties have largely agreed the amount of tax at stake.

the facts

agreed facts

5.             The parties agreed a statement of facts. I mention in passing that there were two sentences in it that were not agreed, but neither party suggested to me that there was anything significant to the appeal in their failure to agree that part of the statement and I cannot see that the disagreement was significant, and in the absence of either relevance or submissions, I do not resolve it.

6.              The agreed facts were that both appellants were companies in the EDF Energy PLC corporate group, by which company they are, indirectly, 100% owned.  The business of the group is energy generation and distribution in the UK.  The group supplies gas to customers over the National Transmission System (‘NTS’) owned by National Grid.  As part of its business, the group has cause to store gas. 

7.             The business of CCS and EDF is the development, construction and operation of gas storage facilities in the UK.    The two companies operate gas storage facilities on adjoining sites in Cheshire.  Between them, the two sites represent 7% of the UK’s gas storage capacity and can hold up to 20% of the UK’s daily delivery capacity.

8.             The site operated by CCS (Hill Top Farm) under a lease comprised, when the site was acquired in 2009, 10 underground cavities filled with brine. In 2014, CCS de-brined three of these cavities, and, at the time of the hearing, was in the process of de-brining two more, to make them suitable for gas storage, at a cost of approximately £40million. The site operated by EDF (Hole House) contained one underground cavity for gas storage when the lease over it was acquired in 2002; in 2002-6, EDF created three further gas cavities by a process of leaching and de-brining at a cost of approximately £j million.

9.             While not a part of the agreed facts, although it seemed agreed by the parties, I find that, in order to use the cavities, CCS and EDF incurred expenditure on boreholes, pipework, pumping and dehydration equipment, and on control mechanisms. All this expenditure has been accepted by HMRC as qualifying for plant and machinery allowances.  The dispute related solely to the cost of leaching some and de-brining the eight cavities.

Findings of disputed facts

Evidence of Mr Christopher Andrew Bebbington

10.         Mr Bebbington was Chief Technical Officer of EDF Energy Plc, and his role involved oversight of the group’s engineering, supply chain and project development for coal, gas and renewables, including oversight of CCS and EDF.  He gave evidence of fact.

11.         He explained that EDF Energy PLC wanted to diversify into gas storage as the UK’s own production of gas declined and it became more dependant on imported gas; doing so would put EDF in a position to maintain a gas supply to customers in the event of interruptions in supply from abroad.  It also would allow it to benefit from price increases in that it could store gas when cheap and sell it when expensive.  It was his evidence that EDF’s business plan for the appellant companies was to profit from arbitrage on gas price volatility.

12.         He also said, and Mr Dorsett agreed, that while the economics favoured investment in fast-cycle gas storage in around 2000 (due to supply and price volatility as explained above), by 2010 there was much less supply and price volatility in the market.

13.         It was his evidence that the two appellants, although owning adjoining sites, to a large extent operated separately.  Each site had its own above-surface plant.  There were pipes which connected the two sites but it was rare from them to be used as the gas in each site would normally be at a different pressure.  The companies did share some staff; in particular both sites were managed by the same people and shared a single team of engineers.

14.         While he was not giving expert evidence, in his oral evidence his explanation of how the two sites operated was very similar to that of the experts.  He appeared to be a reliable witness and I accepted his evidence.

Expert evidence

15.         Mr David Aron was instructed by HMRC.  He was accepted to be an independent expert.  He was managing director of Petroleum Development Consultants Ltd; his background was as a petroleum and chemical engineer with some 40 years’ experience in the oil industry.  He was a fellow of the UK Institution of Chemical Engineers and the UK Energy Institute. He was a member of the UK Institution of Gas Engineers and Managers, and a Founding Member of the European Section of the Society of Petroleum Evaluation Engineers, and a member of the Association of International Petroleum Negotiators.

16.         The appellants did not challenge his status as an expert nor the relevance of his evidence but Mr Codara did make the point that Mr Aron had little direct experience of a gas storage business.  When running his own company, Mr Aron had had experience of putting gas into a storage facility (the Rough offshore field), but his company was the customer of the gas storage facility and not a company whose business was gas storage.

17.         Mr Codara was fairly critical of his evidence, suggesting to me that Mr Aron had guessed answers.  I did not take that impression away. I found him to be a careful witness. I did conclude that Mr Aron’s understanding of what words meant was grounded in the technical or industry usage of the words and would not necessarily reflect the colloquial or general understanding of the word.  This was apparent, for instance, from the disagreement between the experts on what was a pressure vessel and what was a pump.

18.         Mr David Graham Dorsett was instructed by the appellants. He was also accepted to be an independent expert witness.  He was a director of DGD Energy Consultancy Ltd; his background was in mechanical engineering.  He spent 10 years as a senior executive directly involved in the development, construction and operation of the Holford gas storage facility, which stored gas in gas cavities in Cheshire in similar manner to the appellants.  He was a member of Gas Storage Operators Group, the trade association representing the UK gas storage industry and appointed vice-chair of the Board of the Utility Networks Division of the Energy and Utilities Alliance.  He was co-author of the EUA report ‘Gas Storage: securing the future of the UK energy market’.

19.         Mr Dorsett was, clearly, of the two experts, more knowledgeable about gas cavity storage facilities as he had 10 years of direct experience of it.  But this did not really seem to make a difference to their evidence in that both experts, as I have said, were largely agreed on how the sites operated.

20.         Their differences of opinion seemed to be mainly limited to definitions and the meaning of words and phrases, such as what was a pressure vessel and what was a pump.  The experts submitted separate reports, and both submitted further, supplemental reports.  They also agreed a joint statement which set out the main points of disagreement and then both attended the hearing and were cross-examined. 

21.         I accepted both experts’ evidence in large part; it was, as I have said, consistent in essentials about gas storage.  In so far as they disagreed with each other, I have set out below my findings.  First, I set out a summary of what they largely appeared agreed on. 

The gas industry

22.         Mr Aron’s evidence was that the gas industry was comprised of five types of operation:

(1)     Extraction and production.  Gas is extracted and produced from natural occurring gas reservoirs whether with or without petroleum;

(2)     Processing. Gas produced from reservoirs will contain contaminants and be of varying calorific value.  The purpose of processing was to reduce the contaminants in, and ensure a uniform calorific value of the gas so that it met the required standard of gas in the National Transmission Service (‘NTS’).  This included chemical processing (such as ‘washing’ the gas with acid to remove carbon dioxide).

(3)     Transportation/transmission.  Gas was transported through the NTS in large pipes at high pressure direct to large users (such as power stations) or to local gas distribution systems which lowered its pressure to a level suitable for local distribution.

(4)     Distribution.  Local gas distribution systems ultimately delivered gas to end users at just above atmospheric pressure.  This would include intermediate pipes which transported gas at lower pressure than the NTS but higher than would be delivered to consumers.  Gas pressure would be reduced at the junctions between the NTS and intermediate pipes, and the junctions between the intermediate and local distribution pipes.

(5)     Storage.  Storage of gas which had already been processed and which, subject to removal of any contaminants from the storage process, is ready to be put (back) into the NTS.

23.         To a large extent, this evidence was uncontroversial and self-evident.  Mr Codara was unhappy with Mr Aron’s split of the transportation/transmission of gas from the distribution of gas.  While Mr Aron’s evidence certainly seemed to reflect the current set up in the industry with National Grid responsible for ‘transportation’ and various gas companies for the ‘distribution’, as Mr Aron himself said, before privatisation there had been a single nationalised industry which was responsible for delivering the gas from the ‘beach’ (as the experts put it) to the consumer. 

24.         I accept Mr Dorsett was right to say that all the pipes were a single system as all interconnected, but I also accept that Mr Aron was right to say that the gas industry saw transmission (in large pipes at high pressure to gas companies and power stations) as a distinct activity from distribution (to consumers in smaller pipes at lower pressure).  But that does not answer the question of what was meant by ‘distribution’ in the Capital Allowances Act (‘CAA’), which is a question to which I return later. 

25.         I consider the question of whether gas storage was a part of gas distribution/transmission below.

How gas is stored

26.         Mr Aron explained that there were five ways of storing gas, which were in:

(1)     gas holders (also referred to as gasometers) which are now entirely, or almost entirely, disused but a familiar feature in many towns.  They are the distinctive, above ground, large cylindrical containers, which used to rise up and down depending on how much gas they contained, and which are supported by external cross-girders.  They held gas at atmospheric pressure;

(2)      abandoned gas fields, which, being underneath the sea, are completely invisible.  These are beds of porous rock (normally sand), capped by salt rock, to which gas had migrated from shales and been unable to escape.  Once the naturally occurring reservoir of gas had been produced, the depleted field can then be used for gas storage by pumping gas back into the field.

(3)     abandoned acquifers.  Some acquifers, no longer useful for the storage of water, can be converted to store gas.    This is expensive and the UK has had no need to store gas in this way having better alternatives.

(4)     gas caverns.  These are what are in issue in this appeal.  They hold gas at high pressure.  A single cavern can hold about the same amount of gas as 300 gasometers.  They are considered much safer than gasometers as the gas is held underground

(5)     Line pack.  The gas transportation system, in effect the national and intermediate local pipelines, can effectively be used to store gas in that the amount of gas in the pipe system depends on its pressure; increasing the pressure increases the amount of gas and in effect allows gas to be ‘stored’. 

27.         My understanding from the evidence was that there were significant practical differences between the different types of storage.  For instance, removing gas from storage in abandoned gas fields and acquifers would be slow and likely require removal of significant contaminants.  Gas caverns, on the other hand, involved fewer contaminants and allowed ‘fast cycle’ of gas in that the gas could be removed (in relative terms) quickly.

28.         This evidence was also largely uncontroversial; the only area of disagreement appeared to be over whether line pack was used for storage.  Both experts were agreed that the NTS had a minimum and maximum safe operating pressure and so it was possible to ‘store’ gas in the NTS.  Mr Dorsett’s view was that National Grid would not primarily use this ‘line pack’ for storage; Mr Aron’s view was that gas for London was stored in line pack in the intermediate pipes in what he referred to as the distribution system.  It seemed to me that the experts’ views on linepack were really very similar and I agree with Mr Cordora that their  differences on this point were not material to this appeal and I do not resolve them.

Leaching cavities

29.         The experts, Mr Bebbington, and the parties all seemed agreed on how the gas cavities were created, and below is my summary of the process.

30.         In Cheshire, in particular, there was and is naturally occurring halite (salt rock) underground.  Where it is suitably located, it can be solution mined by drilling a borehole into it.  The water slowly dissolves the salt rock.  The borehole is designed to permit the extraction of much of the resulting salty water (brine).  And if the purpose of the operation was to be obtain salt, the salt can then be extracted from the brine with a surface level operation.  Extraction of salt from the ground in this way eventually creates a tear-shaped cavity underground full of brine.  It is a process referred to as ‘leaching’.

31.         The cavities are never emptied, but at the end of the leaching process are left full of brine. The brine exerts pressure on the surrounding rock, whereas an empty cavity would risk fractures in the surrounding rock and ultimately total or partial collapse of the cavern and possible subsidence of the surface.  The cavities are not simply left full of water because any water would react with the halite surrounding the cavity making the cavity even bigger and potentially unstable. (Mr Bebbington mentioned that less careful solution mining in the past had led to unexpected lakes in the Cheshire landscape.) 

32.         British Salt, from whom CCS had indirectly acquired the Hill Top Farm site, had created 10 cavities at the site in this manner in order to obtain salt for its business. However, after acquisition for the purpose of conversion to gas cavities, CCS discovered 3 of the cavities had interconnections which made them unsuitable for use to store gas.  For safety reasons, CCS took the decision not to use those 3 cavities plus the 2 closest to them.  So only 5 of the 10 salt cavities on CCS’ site have been used for gas storage.

33.         As I have said, EDF had earlier acquired the Hole House site with one cavity and created three more cavities in a similar manner to those created for salt mining.   It took about two years to create each gas cavity in this manner.

De-brining cavities

34.         Creating a cavern by means of leaching out the halite results, as I have said, in a tear drop shaped cavern full of brine.  And, as I have said, because of the risk of partial or total collapse, the cavern cannot simply be emptied. So, over a period of about three months, a suitable salt cavern can be converted to gas storage by the slow exchange of the brine with gas via pipes through the borehole. 

35.         At the end of the process, marl (insolubles originally present in the halite) and a small amount of brine will remain at the bottom of the cavity; otherwise the cavity will hold gas.

36.         Once filled with gas, sufficient gas to fill the cavern at a certain minimum pressure must always be left in the chamber to avoid the risk of partial or total collapse of the cavity.  This gas is referred to as ‘cushion gas’.  It cannot be sold into the NTS as it must remain in the cavity.  HMRC accept that the cushion gas in the 8 cavities the subject of the appeal was owned by the appellants and was plant.

The properties of the cavities

37.         The halite in which each cavity was created naturally forms an impervious barrier to the gas.  Therefore, the gas pumped in does not leak out of the cavity into the surrounding rock.  So the cavities did not need to be and were not artificially lined in any way.

38.         There is a maximum pressure of gas which a cavity can safely hold, beyond which any higher pressure would risk damaging the equipment in the wells and fracturing surrounding rocks.  The maximum pressure at which gas can be safely stored in a cavern depends on various matters including the depth of the cavity:  deeper cavities can store gas at higher pressure.

39.         The difference between the amount of gas which may be stored at the highest safe pressure, and the minimum amount of gas which must be stored to maintain integrity (the cushion gas) was known as ‘working gas’.  It was the gas that can be stored in, and removed from,  the cavern.  Obviously, in practice cushion and working gas were all the same gas stored in the cavern.  Referring to gas as ‘cushion gas’ is simply a shorthand method of referring to the minimum amount of gas that had to be left in the cavern for reasons of safety.

40.         It was agreed by the parties that any working gas owned by the appellants was their stock in trade.

Moving gas in and out of caverns

41.         It was accepted that the gas cavities were connected to the NTS.  The cushion and working gas came from the NTS and the working gas would at some point be restored to the NTS. 

42.         On the surface above the cavities, or near them, and connected to each other, the cavities and the NTS by a system of pipework, was plant which comprised:

(1)     Compressors to move the gas in and out of the caverns and from or to the NTS;

(2)     Gas treatment plant.  Its purpose was to remove moisture (hydrate) and to remove other impurities (condensate) accumulated in the gas while in store in the cavities before being returned to the NTS.  The treatment plant could also change the gas temperature to ensure it was within the acceptable temperature parameters for the NTS.

Within the boreholes that connected the cavities to the pipes (and through which gas was introduced and removed) sit valves and temperature and pressure monitors.  Nothing but gas, marl and brine are in the cavity itself.  And, it was accepted, it was not possible for a person to enter a cavity; the only point of access was the borehole in which the pipes sat.

43.         Both experts’ evidence was that the processing carried out when gas from a cavity was returned to the NTS was qualitatively different to that carried on by plant at the ‘beach’ which processed newly produced gas.  Gas freshly produced would have many different sorts of impurities whereas stored gas taken off the NTS only picked up impurities from where stored.  Beach processing would also include chemical processing.  It would alter the calorific value of the gas to ensure it was within acceptable parameters for domestic use (neither too strong nor too weak).  Processing after storage in a gas cavity would not alter the calorific value, which would have been unaffected by the storage.    I accept their evidence on this as this evidence appeared consistent and to make sense:  what it means for the appeal (if anything) I discuss below. 

Free flow of gas from or to a cavity

44.         The experts were agreed that it was possible for gas to free flow from or to the gas cavity.    If the valves between a cavity and the NTS were opened, and the gas in the NTS was at a different pressure to the gas in the cavity, ordinary laws of physics would cause free flow of the gas from the higher pressure area to the lower pressure area.

45.         Everyone appeared agreed that the appellants would, if taking gas from the NTS or putting gas onto the NTS, take advantage of free flow if it was possible to do so.  Free flow cost nothing whereas the alternative was to use the compressors to move the gas and that would be expensive.  I had no factual evidence on how often the appellants actually moved gas by free flow rather than by using the compressors.

46.         Mr Dorsett’s expert evidence was that, as the NTS was normally at a higher pressure than either of the appellants’ cavities, any free flow would occur on gas entering the cavern rather than leaving it.  Mr Aron’s evidence was that free flow would require a significant difference in pressure between the NTS and the cavities because the pressure differential had to be strong enough to draw the gas through the appellants’ pipes and processing plant.  It seems to me that it  follows logically that this loss of pressure due to ‘resistance’ would be more significant on leaving rather than entering the cavities, as the gas was only processed on leaving,  and so, again, consistently with what Mr Dorsett said, suggested free flow was more likely to occur, if at all, on entering rather than on leaving the cavities the subject of the appeal.

47.         Mr Dorsett did not give an opinion on how often free flow would occur; Mr Aron did not know either, but was inclined to think it infrequent.  Mr Aron also made the point that, in his view, free flow could not be planned in advance as it depended (in part) on the pressure of the NTS which would be out of the appellants’ control.  Mr Dorsett pointed out that the appellants had some control over the pressure in the cavity, so he thought it too simplistic to say free flow was entirely out of the appellants’ control.

48.         I find free flow was possible, but could not be planned in advance as it depended at least in part on factors beyond the appellants’ control and was most likely to be feasible with gas entering the cavity.  The appellants would most likely have to turn on the compressors if moving gas from a cavity back to the NTS. 

Gas temperature

49.         Mr Dorsett’s evidence was that the temperature of the gas on entering the cavity would be influenced by its temperature in the NTS, the temperature of the gas already in cavity, the speed at which the gas was piped into the cavity, its pressure and the temperature of the surrounding rock.

50.         I do not think Mr Aron disagreed with this but it was his view that the most significant influence on the temperature of the gas in the cavity would be its pressure.  Mr Aron accepted that he lacked experience to know what the temperature of the gas in the cavity would be.  His view was that the pressure of the gas was much more significant to the appellants than its temperature.  He did accept that its temperature and pressure were related in that increased temperature increased pressure and for that reason it was better to pipe in the gas slowly so that the temperature rise was lower.  Nevertheless, his evidence that temperature was less significant than pressure made sense as pressure related directly to how much gas could be stored. 

51.         I find that while the temperature of the gas clearly mattered to the appellants (as it related to pressure, how hydrated the gas became, and because gas had to be within certain temperature parameters to be restored to the NTS), and while temperature of the stored gas would be affected by the temperature of the rock in which it was stored, there was no evidence from which I could conclude that the appellants used the cavities to influence the temperature of the gas to any significant extent.  On the contrary, my understanding of what  Mr Dorsett said was that it was the surface plant (and not the cavities) which was used to make any significant alteration to the temperature of the gas either on entry to or exit from the cavities.  Moreover, it was a necessary implication of what he said that any need to alter the temperature of the gas arose from compressing it and/or storing it and was an incident of the storage and certainly not the purpose of the storage.


For what did the appellants use the cavities?

52.         It seems that the appellants did not appreciate until the hearing was underway that there might be a dispute over the use to which the cavities were put. 

53.         Mr Aron’s expert evidence was that CCS appeared to be fast-cycling gas (because the balance of gas held regularly changed) and therefore was using its gas cavities for trading of gas.  He was unable to say on the data available for what EDF used its cavities.   Mr Dorsett referred to the appellants being in the gas storage service industry.

54.         The factual evidence on this in large part comprised the accounts as Mr Bebbington was not really asked.  He had agreed with counsel that CCS’ business was gas storage but that did not indicate who owned the gas stored nor how fast was the turnover.

55.          The accounts for CCS showed that, by 2012 and thereafter, CCS owned significant amounts of working gas, indicating that the fast cycling identified by Mr Aron was on CCS’ own account. 

56.         The accounts for EDF, however, as Mr Cordara accepted, showed that EDF did not own the gas it stored for the years at issue in this appeal.  Its income came from payments from another group company for storing the gas of which that company was owner.  The accounts also stated that its income was impaired because gas price volatility was reduced; Mr Cordara inferred from this that EDF also fast-cycled gas, albeit gas belonging to another group entity. 

57.         Ms Nathan emphasised her view that there was insufficient evidence on which I could reach the conclusion that the appellants’ fast-cycled gas or that CCS owned the gas in its cavities, but I do not agree.  I find that the facilities, on the basis of the evidence before me, more likely than not were used for fast cycling storage of gas, although in the case of EDF, it did not own the gas that was fast cycled, but was paid by the owner(s) for moving and storing its gas.  In other words, the CCS more likely than not intended to make profit from gas price volatility by trading gas which it owned; EDF more likely than not intended to make profit by charging customers for using its storage cavities so that its customers could make profits from gas price volatility.

Were the appellants in the business of gas distribution or processing?

58.         The appellants put the case that their business of storing gas meant that they were in the business of gas distribution and gas processing. 

59.         So far as gas distribution was concerned, they saw storage of gas as an integral part of distribution.  When gas in the cavity was returned to the NTS, it was, they said, distributed.  Moreover, the gas was processed to remove water and condensates before it was returned to the NTS.

60.         However, while I accept that as a matter of fact the appellants did process the gas before returning it to the NTS, I do not accept that they were in the business of processing gas.  The evidence of both experts was clear that the processing was merely a necessary incident of storing the gas:  to be permitted to return gas to the NTS from the cavities it had to be processed to restore it to the standard of purity it had when it was removed.

61.         Mr Dorsett said that gas storage played a role in the gas system, but nevertheless I understood that both experts were agreed that gas storage in cavities was unnecessary to gas distribution in the sense that there could be a gas distribution network without gas cavities.   Moreover, I find that returning the gas to the NTS was merely a necessary incident of the appellants removing the gas from the NTS in the first place in order to make to make profits from storing gas.  In conclusion, I do not accept that the appellants were in the business of gas distribution either; their business was to store gas, even if only for a short period.  Returning it to the network made the gas available for distribution but distributing it was not their business.

62.         I find that the business of the appellants’ was gas storage and not distribution or processing of gas.

Areas of disagreement

63.         The experts, as I said, compiled a list of issues on which they did not agree and they were as follows.

Were gasometers tanks?

64.         Mr Aron’s view was that gasholders were not tanks.  He relied on a study of gasometers which made a distinction between the gas holder and the water tanks contained within them as an essential part of their mechanism as they prevented the gas escaping.  The basis of his view that the report was right not to consider gasometers to be tanks seemed to be because they were part of an active system storing gas at night and releasing it in the day: he thought the report correct to describe them as gas holders.  He accepted that tanks could have floating tops and so his view that gasometers were not tanks was not based on the gasometers’ ability to change size.

65.         Mr Dorsett disagreed with him on the basis that the dictionary definition of a tank was ‘a large container for holding liquid or gas’ and therefore a gasometer, as a large container for holding gas, was, in his view, a tank.  Neither the report relied on by Mr Aron nor he fact that a gasometer had telescopic sides, altered his view it was a tank.

66.         This disagreement, as with others, indicated to me that the difference in view between the experts was because Mr Dorsett tended to view things with the same function as the same technical thing; Mr Aron was concerned to use terms as he understood they would be used in the industry in which he worked, and that to him meant he had to consider method of construction as well as function.  I did not think that either expert was so much right or wrong but just having a differing view point.

67.         From the point of view of this decision, I did not need to resolve whether, in the gas industry, it was technically correct to describe a gasometer as a tank; I did need to decide what ‘tank’ meant as a matter of law and that was not something on which the experts could, or purported to, give evidence.  I address it below.

68.         Both experts did appear agreed that a tank would be artificially constructed and would be able to hold its shape when empty (at least as long as internal and external pressures were equalised).

Were cavities tanks?

69.         Mr Aron did not think it would be technically correct to describe a gas cavity as a tank. HMRC’s view was that a gas cavity, although artificially created, was just a hole in the ground and at risk of collapse if not filled.  It was suggested to Mr Dorsett that the cavities were not tanks because they were at risk of collapse if emptied of gas or brine. He pointed out that above ground tanks would also be at risk of collapse if emptied of air.  He considered that because they had the same function as above ground tanks, the underground cavities were tanks.

70.         I consider that whether the cavities were tanks within the meaning of the CAA was a question of law and one which I will determine below.

Did the gas caverns operate as pumps?

71.         Mr Dorsett’s opinion was that gas cavities were able to act as a pump. Mr Aron did not agree.  This led to a discussion between the expert on what a pump actually was.

72.         Mr Aron’s view was that the gas cavities could not operate as a pump in moving gas because a pump did not move gas.  Compressors moved gas.  While Mr Dorsett accepted that compressors only moved gas (because only a gas and not a liquid could be compressed), he considered a pump could be used for any fluid, either gas or liquid (but not both at the same time).

73.         Mr Aron explained his view on the basis that, while both compressors and pumps are rotating equipment (with motors and impellers), in industry compressors used to move gas and pumps are used to move liquids.  Liquid in a compressor could cause catastrophic failure, gas in a pump would mean it was inoperable. Compressors tended to be very large and fast while pumps tended to be smaller and slower.

74.         Mr Dorsett accepted that, in some industries, it was convention to use pumps only for liquids, but did not agree that it was always wrong to describe equipment which actually moved gas as a pump.  As an example, he referred to a bicycle pump.  Mr Codara also pointed out that Mr Aron himself had referred to the equipment which put the gas smell into the gas as a ‘small pump’ even though recognising it pumped gas.  Mr Aron’s view was that colloquial terminology could be at odds with technical terminology and it was not technically correct to describe rotating equipment which moved gas as a pump.  A bicycle pump was technically a piston compressor.

75.         The discussion was very interesting but, it seemed to me, quite irrelevant other than to reinforce the view set out at §20 on the experts’ differing approaches to terminology. What was clear was that both pumps and compressors had moving parts and the gas cavities did not.  So when Mr Dorsett said that cavities acted as a pump, he did not mean that they actually had any moving parts.  A cavity was not actually a pump or compressor.  What he meant was that the cavity could have the same effect as a pump/compressor in that (he was saying) it could move gas. 

76.         To some extent, what he was saying was no more than what Mr Aron agreed with him on, which was that the pressure differential between the gas in the cavity and the gas in the NTS meant that, when the valves between them were open, there could be free flow as gas would naturally move from higher pressure to lower pressure areas.  However, Mr Dorsett was also saying that the cavity caused this effect, and I did not think Mr Aron agreed with him on that.  Mr Aron’s view was that the cause of the free flow was the pressure differential.  The cavity did not even create the pressure differential because the cavity was a passive hole in the ground.  The pressure differential was caused by the pressure of gas stored in the cavity versus the pressure of gas in the NTS.  The cavity was only responsible for the pressure of the gas in the cavity in the sense that the rock in which the cavity was formed was impervious to gas and so prevented it leaking away and the pressure reducing:  indeed it might be said that as these cavities were normally at a lower pressure than the NTS, the halite actually reduced the possibility of free flow by maintaining the pressure of the gas and not allowing it to reduce further.  (This could not be said of the cavities with which Mr Dorsett spend 10 years of his career at Holcomb as they were deeper and typically stored gas at a higher pressure than the NTS).

77.         My conclusion was that it was only possible in the loosest sense to say that gas cavities acted as pumps:  while free flow was possible where there was sufficient pressure differential between the cavities and the NTS, the movement of the gas was caused by the pressure differential which depended on the pressure of the gas held by both the cavity and the NTS.  The cavity was a part, but only a part, of this system in that it was where gas was stored.  The entire system might be said to act as a substitute for a pump/compressor, or at least to make it (on some occasions) unnecessary to turn on the compressors, but, I consider, the cavity by itself was not a pump and did not, by itself, act as a pump. 

Do the gas caverns operate as pressure vessels?

78.         Mr Dorsett said that the caverns operated as pressure vessels as they permitted gas to be stored at greater than atmospheric pressure.  Mr Aron agreed that the caverns permitted gas to be stored at pressure but did not agree that that made them pressure vessels.

79.         Mr Aron’s starting position was that the cavities were not pressure vessels as they were not covered by the EU Directive on pressure vessels which was intended to ensure all pressure equipment in the EU was made to same standard.   Mr Dorsett in his supplemental report pointed out that the Directive was not (in his opinion) intended to cover gas cavities and so it was wrong to draw the conclusion that a gas cavity was therefore necessarily not a pressure vessel.  Mr Aron originally accepted this but in cross examination appeared to have forgotten he had done so as he reverted to his view that the caverns were not pressure vessels as they were not covered by the Directive.

80.         However, it was also apparent from what he said, that Mr Aron could not accept that the cavities were pressure vessels as he could not accept that a hole in the ground could be described as a vessel.    This view seemed aligned with his view that the cavities, as holes in the ground, were without manufactured sides and could not be described as tanks or containers nor as pieces of equipment.  (I note at one point in his evidence he did not notice that Mr Cordara described a cavity as a ‘container’ but focussed instead on the question which raised a different issue:  when asked directly, he said a cavity was not a container.)

81.         I do not have to decide whether the cavities are pressure vessels as a matter of law; it is not a term used in the part of the CAA I am called upon to consider.  It is clear that the cavities were designed to, and did, hold gas at pressure.  It is also clear that, although artificially constructed, they did not have sides which were manufactured.  The sides to the cavity were naturally occurring rock.  Whether this is material to any of the issues I must decide under the CAA, I will deal with below.


the law

82.         It is a basic tenet of tax law that capital expenditure is not deductible from trading profits when calculating liability to tax.  The limited exception to this is that some deductions can be claimed for certain capital expenditure.  The legislation providing for this is contained in the Capital Allowances Act 2001.

83.         That provides as follows:

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.

84.         HMRC accepted that the expenditure on leaching and de-brining was incurred in respect of a qualifying activity, as they accepted that it was incurred for the purposes of the appellants’ trade, and trade was a qualifying activity.

85.         Originally, it appeared that HMRC did not accept that the appellants came to own the cavities as a result of incurring the expenditure on them.  The appellant could only speculate on why HMRC held this view, as although the appellants were only tenants of the sites under which the cavities were created, since 1985 the law has been that in some circumstances a tenant can obtain capital allowances for expenditure on leasehold property.  HMRC did not pursue the point in the hearing and were in no position to do so in any event having not made their case clear on it.  For the purpose of this appeal, I take it that condition 11(4)(b) is to be treated as satisfied.

86.         The dispute the subject of the appeal was under s 11(4)(a).  HMRC did not accept that the expenditure was expenditure ‘on the provision of plant and machinery’.   As it was not suggested the expenditure was on ‘machinery’, the dispute was whether the expenditure was on ‘plant’. 

Order in which law should be considered.

87.         There is a great deal of case law on the meaning of ‘plant’ and the parties did not agree whether or not under this case law the cavities would be seen as plant.  In addition, there is now a legislative code which excludes from the benefit of allowances certain items which would otherwise qualify as plant and again the parties were not agreed whether it applied to the cavities.  And the parties were not even agreed on the order in which I should approach the resolution of these disputes:  should I consider whether the cavities were ‘plant’ before or after I considered whether expenditure on them was expressly excluded by provisions of the CAA from being eligible for capital allowances, or the other way around? 

88.         Mr Codara’s view was that it was logical to address whether the cavities were plant under common law before looking at statutory exclusions.  Judge Poole had agreed with the appellant on this in SSE Generation [2018] UKFTT 146 at §§26-28 on the basis it was logical first to look at whether thing was plant before looking to see if it was excluded from benefit of the allowances.

89.         HMRCs position in that case and this was that allowances were statutory so it was more sensible to look at the exclusions first before considering whether it was plant, particularly as Ms Nathan appeared to view the case-law on plant as more convoluted than the statutory exclusions.  She noted that the SPC had chosen to look at the exclusions first in the case of Anchor International Ltd, [2003] UKSC SPC00354, although I find on appeal ([2005] STC 411) the High Court did not say which was the proper course for a first instance court to follow. 

90.         Ms Nathan relied on the explanatory notes to the exclusions and the Hansard report of the Parliamentary debate on them.  Her position was that these made it clear that the purpose of the statutory exclusions was to prevent further erosion (which had occurred by judicial interpretation of the meaning of ‘plant’) of the boundary between plant on the one hand (entitled to allowances) and buildings and structures on the other hand (which should not obtain allowances due to their long life).  Mr Cordara complained that Hansard should not be referred to as there was no relevant ambiguity in the legislation which needed elucidation.

91.         I did not think that this was a case where the answer to the question of whether the cavities were plant, or the question of whether the cavities were excluded by statute, were so clear that I would decide the case without considering both questions.  And as neither party suggested, quite rightly, that the order in which I considered them would have any effect on the outcome, the order in which I considered them did not seem to matter and this dispute was therefore rather anodyne.

92.         It seemed to me that the first stage was to identify what it was that the appellant claimed was plant; and then, whether I considered the exclusions first or the meaning of plant first, did not matter as it would not affect the outcome.  Clearly, I had to take one issue before the other. 

 What is the alleged plant?

93.         As I understood it, the claimed plant was the cavities.  The appellant wanted capital allowances for the costs of leaching and de-brining the cavities.  While there was a borehole with pipes and equipment within the pipes, these were already accepted to be plant and no issue arose in respect of them.  The question was simply whether the cavities themselves were plant so that the costs of creating them were entitled to capital allowances.

94.         In so far there was a distinction in capital allowances cases between what is referred to as the piecemeal and entirety approach, neither party was advocating an entireties approach.  The question was whether the costs of de-brining and leaching, or either of them, were entitled to capital allowances.

Were the cavities plant?

The case law on plant

95.         As I have said, the definition of plant is in case-law.  Plant is to be distinguished from stock-in-trade and from premises:

In its ordinary sense….[plant] includes whatever apparatus is used by a business man for carrying on his business – not his stock-in-trade which he buys or makes for sale; but all goods and chattels, fixed or moveable, live or dead, which he keeps for permanent employment in his business.

Per Lindley LJ, Yarmouth v France (1887) 19 QBD 647

There is a well-established distinction, in general terms, between the premises in which the business is carried on and the plant with which the business is carried on.  The premises are not plant.

 Per Fox LJ Wimpy International Ltd v Warland [1989] STC 273, 279

96.         The cases establish that the distinction between plant and stock-in-trade and plant and premises is one of function in the taxpayer’s business:  does the thing function as premises or as plant or as stock-in-trade?

In the end each case must be resolved, in my opinion, by considering carefully the nature of the particular trade being carried on, and the relation of the expenditure to the promotion of the trade.

Per Lord Wilberforce in IRC V Scottish and Newcastle Breweries [1982] 2 AER 230


The question in each case is….does the item function as premises or plant?  To answer this may involve deciding whether it is more appropriate to describe the item as apparatus for carrying on the business or as the premises in or upon which the business is conducted…..

Per Peter Gibson LJ in Attwood v Anduff Car Wash Ltd [1997] STC 1167 at 1177

Plant and stock in trade

97.         HMRC did not suggest that the cavities were stock in trade.  It was accepted that the cushion gas was plant and the working gas was stock in trade. 

Premises or plant?

98.         HMRC’s position was that the gas cavities were the housing, setting or premises from which the appellants’ business was carried on and was not the plant with which the business was carried on.  HMRC compared the cavities to warehouses where goods are stored; warehouses are premises and not plant.  They safely house the products which the business exists to sell.  HMRC considered the appellants’ business to be one of gas storage and did not accept that the gas cavities should be compared to gasometers.

99.         The appellants said the cavities were not passive as they stored gas under pressure and in some circumstances that pressure was used to move the gas to the NTS. In any event, the Appellants maintained that cavities could be plant even if their function was entirely passive citing Jarrold v John Good 40 TC 681.

Quarantine Kennels

100.     So what is the test to distinguish plant and premises?  In Carr v Sayer [1992] STC 396 the High Court tried to draw principles together.  The principles identified in that case were:

(1) ‘plant carries with it a connotation of equipment or apparatus, either fixed or unfixed’

(2) ‘”machinery or plant” is apt to include equipment of any size.  If fixed, a large piece of equipment may readily be described as a structure, but that by itself does not take the equipment outside the range of what would normally be regarded as plant.’

(3) ‘equipment does not cease to be plant merely because it also discharges an additional function, such a providing the place in which the business is carried out’

(4) Buildings…do not cease to be buildings and become plant simply because they are purpose-built for a particular trading activity

(5) One of the functions of a building is to provide shelter and security for people using it and for goods inside it…a building does not partake of the character of plant simply, for example, because it is used for storage by a trader carrying on a storage business. This remains so even if the building has been built as a specially secure building for use as a safe-deposit business….’

Per Sir Donald Nicholls VC

101.     In that case, the Court decided that purpose built fixed quarantine kennels for cats and dogs were not plant because it was where and in which the business of the appellant was carried on.  The court accepted that temporary, moveable kennels would be plant.

102.     Mr Cordara said the Tribunal should use care when considering principle (5) as a safe deposit is treated as plant under the exclusions, to which I refer below.  However, I take account of the fact that the Court of Appeal approved the decision in later case of Attwood v Anduff Car Wash Ltd [1997] STC 1167, saying:

‘Yet the premises test was not satisfied because the kennels performed a typical premises function, providing shelter’

Car wash

103.     The Anduff decision concerned a car wash site where about 10% of the site by area was the wash hall where the car-washing activity took place while the rest of the site was for moving and parking cars, and vacuuming parked cars.  The entire site viewed as a whole was found to function as premises.  Even the wash-hall by itself was found not to be plant:  while it contained plant (eg the washing equipment), and itself functioned in some way as plant (the building was designed to collect water for re-cycling), to the greater extent its function was as premises as it provided protection from the elements for the plant and machinery and workers, it provided noise reduction and a place to work (including a lobby and WC).

104.     So the test is one of function and it can be difficult to apply where the thing in question has both plant-like and premises-like functions.  Mr Codara’s view was that if a thing had any plant-like function, it was plant; it was only if it was ‘mere premises’ and without a plant-like function that it would not be plant.  That view does not seem consistent with Attwood v Anduff where the wash-hall did have a plant-like function (the recycling of water) but it was insufficient to outweigh its premises-like nature of providing shelter.

Dry Dock

105.     The appellant relied on a leading but older case of IRC v Barclay Curle (1969) 45 TC 221 which decided that the work of excavation and concreting of a dry dock was allowable as the dry dock itself was plant.  It might be thought that the dry dock, which comprised the entire premises on which the taxpayer operated its business of ship repair, was premises as the car wash site was later found to be  in Attwood v Anduff, as it provided the place where the work was carried out, which is a premises function. 

106.     However, the dry dock also had plant-like functions, firstly because its function was to hold the ships in place while they were worked on; but secondly and even more significantly, it seems to me, was its critical ability of isolating the ship from the water.  Its function was compared to that of a crane hoisting a boat out of water for repairs; a crane removed the boat from the water: the dry dock removed the water from around the ship.

107.     The Lords ruled that those plant-like functions made the entire dry dock plant.  But while the Lords said buildings and structures could be plant if they had plant-like functions, there is nothing in their decision which implied that any plant- like function would automatically make any building and structure plant.  And the later case of Anduff shows that this is not so. I conclude that it is a matter of degree where premises have both plant and premises like functions: the question must be which of the functions predominates.  With Barclay Curle, the plant-like function of the elimination of the water was critical to the business.  In Attwood v Anduff, the premises-like function of the wash hall predominated over its re-cycling of water.

The water tower

108.     Another case concerning the border between premises and plant to which I was referred was Margrett v The Lowestoft Water and Gas Company (1935) TC 481.  This concerned an above-ground water tank supported by a structure.  The tank, by harnessing gravity, was used to increase the pressure of the water supply in the eponymous town.  The decision itself is not relevant because the House of Lords in Barclay Curle said it was wrongly decided.  The significance in the case is that the Lords said that the water tower should have been treated as plant because it had a function beyond storing water. It was a structure and it provided shelter and containment of the water, but its function was to increase water pressure and that made it plant.

A swimming pool

109.     Another case which considered the borderline between structures which were premises and those which were plant was Cooke v Beach Station Caravans [1974] STC 402.  The taxpayer in that case claimed that a swimming pool was plant so that he was entitled to allowances on the cost of excavation, terracing, and construction of pool, including concreting and lining it.  HMRC’s case was that the pool was merely setting for the appellant’s business of a holiday park. 

110.     Megarry J referred to Barclay Curle in his decision and was clearly aware of Lord Reid’s statement that:

 …a lock which impounds water is not plant although a trader uses it as the source of the water he needs.  And a dam is generally simply an improvement of the lock giving a better supply, But I could imagine circumstances in which a dam would be such an integral part of the means required for a trading operation that it should be regarded as plant….

He concluded that the swimming pool was more than just a dam holding water; it was an active part of the business in that it was there to provide entertainment for the holidaymakers.  The pool might be a passive reservoir of water in one sense, but it was for active use in that particular business.  It was not merely storing water.

Grain silos

111.     I see the case of Schofield [1975] STC 353, to which I was often referred in the hearing, as being in much the same vein.  In that case, silos for short-term storage of grain were held to be plant.  The silos were said to be in the nature of a tool in the taxpayer’s business of grain importation:  they were designed to allow quick discharge of the grain (by use of gravity) to the trader’s buyers.  The silos were apparatus with which the trade of grain importation was carried out. 

Underground sub-station

112.     I was referred to other cases. Bradley v London electricity plc [1996] STC 450 concerned a large underground substation which housed transformers distributing electricity.  Features of the sub-station, such as the transformers, performed a plant-like function in the appellant’s business of electricity supply and the substation was designed specifically to house them; the substation as a whole was found to function as premises for the trading activity.  The court made the point that the silos in Schofield was not mere shelter but were there for the fast and convenient discharge of grain.


113.     In the Wimpy International Ltd v Warland [1989] STC 273 case the question was whether the internal decorations of cafes were plant.  The Court said that the test was not whether the items were fixtures, although the more affixed the more likely to be premises rather than plant. The Court accepted premises themselves could be (in rare cases) plant.   Lloyd LJ said there was no single test but suggested it would be relevant if the item was housing for business or whether it had an extra function. 


114.       Mr Cordara’s position was that I should accept 4 principles from all these cases and those were that

(1)     Stock in trade is not plant;

(2)     Plant is apparatus used in business;

(3)     Mere setting in which business is carried on is not plant;

(4)     Setting can be plant if it has plant like function.

115.     Ms Nathan accepted propositions (1) and (2) but not (3) and (4).  She said that there was no test of ‘mere setting’; the tribunal should just look at function and decide if its function was that of setting or of plant. I thought that both counsel were really saying much the same thing:  if the main function of premises was as premises, it was not plant; if it was premises or setting, but its primary or significant function was as plant, it would be one of those rare cases where the premises was plant.

Cases on storage

116.     The cases raise the question about structures or equipment used for storage.  It is clear that the case law establishes that providing shelter and protection to plant or to stock in trade or to the taxpayer’s customer’s possessions (see Carr v Sayer) is a premises function, at least where immovable property is concerned.  Moveable storage equipment appears to be plant:  see Carr v Sayer.

117.     Such an analysis makes sense of the various statements about the storage of water.  In Barclay Curle, mere dams holding water, which are clearly premises, were said to have the premises-like function of storing (ie retaining) water.  However, water stored for a purpose other than retention for later use, such as in Magrett, where the water was stored in a tower to increase water pressure in the system, was stored for a plant-like purpose. It was not stored for a purpose that was a premises-like function of protection and retention.  And in my view, this was why the Lords in Barclay Curle said it was wrongly decided that the water tower was not plant.

118.     In Schofield, Lowry CJ commented that

‘one can see how essential are the silos, in contrast with mere warehouses, to the working of the business’. 

Moreover, it seemed important in Schofield that the finding was that the taxpayer’s business was of grain importing (in which the storage of grain was a trifling element) and not the business of grain storage:  it was the speed of input and output into the silos which were important and not their capacity to protect and contain. 

119.     It is not a question of whether the item is essential to the business:  premises can be essential to the business, such as in Carr v Sayer, but that does not make the premises plant.  Premises are not plant, it seems to me, unless its plant like function is more significant to the taxpayer’s business than its premises–like function.

120.     The appellants also relied on the case of Union Cold Storage (1939) 22 TC 547 although they did not refer me to any passages from it.  It was their position that it was decided that cold stores were plant because, despite being premises, they performed a plant-like function of making and keeping food cold while in store.  Cold rooms preserved stored food.  However, as in Schofield, the taxpayer was not in the business of storing food:  it was an importer whose business was to sell food.  So it seems to me the cold room better enabled the taxpayer to perform its function of selling food fit for human consumption.

Were the gas cavities plant?

121.     The cavities were essential, indeed central, to the appellants’ business, but that alone would not make them plant.  Premises and plant can both be essential.

122.     The gas cavities were clearly part of the premises from which the appellants carried on their businesses.  They were not merely fixed to the land, they were a part of it, like an underground reservoir.

123.     That did not rule out the possibility that they were plant.  The authorities were clear that ‘holes in the ground’ could be plant:  Barclay Curle and Beach Station Caravans both concerned holes in the ground.  The more recent case of SSE Generation also concerned underground construction. Whether the cavities were plant depended on whether they functioned as premises or whether they functioned as plant. 

124.     The appellants’ business was, I have found, short-term storage of gas.   The gas was fast-cycled in that it moved in and out of the cavities in the short term.   EDF’s business at the time was to charge other entities for this short-term storage; CCS’s business aim was to profit from storing gas by buying the gas at one price and selling it shortly afterwards at a higher price.

125.      What was the function of the cavities in this business?  Their function was to store gas taken from the NTS safely (so that it did not dissipate) and in a condition that would allow it to be returned to the NTS (or at least in a state from which the appellants’ processing plant would be able to restore it to a condition suitable to be returned to the NTS).

126.     The appellant sought to compare the cavities to the silos in Schofield and the water tower in Magrett both of which had a function beyond that of storage; both used gravity, the silos to enable fast discharge and the water tower to increase water pressure.  The appellants saw the cavities as using pressure to move gas.  They saw them as equivalent to pumps/compressors, using natural forces (as in Schofield and Magrett) in their plant-like function.

127.     However, I think it clear (particularly from Attwood) that a plant-like function does not necessarily make premises plant, in circumstances where the premises also functions as premises. It is a matter of degree.  I am prepared to accept that the cavities did have a plant like function similar to that of a pump/compressor in that the manner of construction (a large hole in the ground connected by pipes to the NTS) meant that when the pressures were right and the valves open, gas would free flow to or from the cavity.  However, I accept the evidence that that was an incident of the construction and not the reason they were constructed in that manner. The main reason gas was stored in salt cavities, as I understood it, was that that gas cavities were a safe method of storing very large amounts of gas in a way that would enable it to be fast-cycled, and thus opened up the possibility of arbitrage on gas prices.  I had no evidence that their plant-like ability to act like a pump/compressor was essential:  on the contrary, the evidence which I had was that the appellants had little control over when they would be able to free flow gas and they had compressors to move the gas when it was not possible.  From the evidence I had I was unable to conclude how often free flow took place and the appellants have therefore failed to show that it was common let alone a main function of the cavities.

128.     In contrast, while the water tower stored water, its purpose was to increase pressure; the silos stored grain but their purpose was fast discharge.  Here the purpose of the cavities was to store gas and not to free flow it.  Where premises have a function of storage, that is a premises-like function and not a plant-like function.

129.     I did not understand the appellants to be arguing that a function of fast-cycle storage was a plant-like function.  Schofield might suggest it was, as the grain was put into and discharged from the silos in short order,  but I do not think that was the ratio of the case.  In Schofield the taxpayer’s business was distribution and not storage.  I consider that in this appeal, the appellant’s business was storage, for however short a term.  They were not, unlike in Schofield or in the case involving a cold store, buying and importing for immediate sale on arrival, but deliberately storing the product (in this case, gas)  in order to profit from a price increase while the gas was stored (in the case of CCS) or to profit by allowing another company to profit from a price increase while the gas was stored (in the case of EDF).

130.     So the premises-like function of shelter and containment was, I find, the significant and predominant function of the cavities.  It makes no difference that there was a fast turnover as the function was storage.

131.     The appellants pointed out that a cold room was premises with a storage function which was plant.  But, as I understand it, that was because the cold room’s main function was to reduce the temperature of what was stored.  While the evidence was that the cavity itself might influence the temperature of the gas, it was not the main influence and, in any event, altering the temperature of the gas was clearly not a function of the cavity. 

132.     Rather, I understood the appellants relied on this case for the proposition that the function of the cavity was not merely to store gas but to store gas at high pressure, and the cavities should be compared to a cold room which allowed storage at low temperature.  The cavities were clearly constructed as they were in a layer of halite which  enabled high pressure storage of gas.  Whether or not technically a pressure vessel, they functioned as a pressure vessel and therefore, said the appellants, the cavities had a plant-like function.

133.     But I consider that the purpose of the high pressure was simply to store more gas. So, the ability of the cavities to store gas at high pressure was a premises like function as it meant that the cavities were just very good premises for storing gas.  Merely being purpose built and very good at performing their premises-like function of storage did not make the cavities plant. 

134.     My conclusion is that the cavities were not plant. That is sufficient to dismiss the appeal. Nevertheless, bearing in mind that this case might go higher and that I had full argument, I will set out my conclusions on the second part of the appeal as well.

Outline of Exclusions from caa

135.     Dealing with exclusions from capital allowances,  s 22 CAA provided:

Section 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 set out in paragraphs 1-6, various types of excluded structure/asset (including tunnels, bridges, dams, reservoirs, dikes, canals, hard standing and docks).  HMRC accepted that the gas cavities did not fall anywhere in paragraphs 1-6.  However, at paragraph 7, List B included within the scope of the excluded structures:

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

(a) …not applicable….

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

(c)…not applicable…..

136.     In summary, putting aside the question whether the gas cavities were ‘plant or machinery’, if they were a structure, they were excluded by List B paragraph 7 unless they were a structure in use for the purposes of an undertaking for the extraction, production, processing or distribution of gas.   HMRC’s position was that the gas cavities, if a structure,  were a structure which was not for the purposes of such an undertaking.

137.     In the alternative, it was HMRC’s case that the expenditure on the gas cavities was expenditure on works involving the alteration of land and therefore excluded by s 22(1)(b).   I will deal with s 22(1) in detail below but at present continue set out an outline of the legislation. 

138.     The other  relevant subsections of 22 CAA were:

(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 s 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.

139.     S 22 CAA also provided, in (4), that it was subject to s 23 CAA.  S 23(3) CAA provided that any expenditure on an item in List C would be unaffected by the provisions of s 22.  List C therefore comprised further exceptions to all the exclusions of s 22, including List B.  The items in List C potentially relevant to this appeal were as follows:

4.  Manufacturing or processing equipment; storage equipment (including cold rooms); display equipment; and counters, checkouts and similar equipment.

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

28. the provision of –

(a) silos provided for temporary storage, or

(b) storage tanks

140.     I will take them out of order as it is more logically to consider the items relating to storage and then the question of alteration of land.  First, I will consider general questions of how the statutory provisions should be understood.

Statutory interpretation

141.     The parties were agreed that statutory interpretation should be purposive but did not agree on what that meant.  Ms Nathan relied on UBS [2016] UKSC 13 and Barclays Mercantile  [2004] UKHL 51 which had cited Lord Wilberforce in Ramsay where he said:

 ‘tax is generally imposed by reference to economic activities or transactions which exist in the real world’

She also relied on [61] of UBS where Lord Reed said:

‘the modern approach to statutory construction is to have regard to the purpose of a particular provision and interpret its language, so far as possible, in the way which best gives effect to that purpose.’

142.     It was her position, firstly, that such purposive construction applied in capital allowance cases as in any other tax case (and cited J D Weatherspoon plc [2012] UKUT 42 (TCC) at [54]: ‘We are required to apply tax law purposively’).  Secondly, it was her point that words should be given real-world meanings. She thought that the Tribunal should understand words ‘processing’ and ‘distribution’ used in respect of gas undertakings in the way Mr Aron understood them as he was an expert in the gas industry.

143.     Mr Cordara did not agree. Apart from the fact that he did not accept Mr Aron’s evidence on this point, he also pointed out that in Britax [2002] EWCA Civ 806, where counsel had suggested that the word ‘hiring’ used in legislation should be given the meaning it would be given in a commercial context, which would not include finance leasing,  the Court of Appeal had not agreed.  The Court said:

  [68] ‘where [statutory] language is capable of bearing more than one meaning, the court will look to construe it in a way which accords with ‘the purpose and the spirit of the legislation’. 

144.     While Ms Nathan said she was uncertain that I should put too much weight on what was said in Brittax which was before cases such as UBS, it seems to me that what was said here was quite uncontroversial:  a purposive interpretation means looking at the context to decide what Parliament intended by the words they used.  So it is possible that they used a word in a technical sense but whether they did so depends on the context.  I will go on to consider what Parliament actually meant by the words they used in the relevant parts of CAA.

Should items on Lists B and C have a functional interpretation?

145.     HMRC’s case was that Lists B and C should be read descriptively:  was the item in question described by anything on List B or C?  The appellants’ case was that the question was whether the function of the thing in question was the same function as anything described on the list.  Their rationale for this view was that, especially with regards List C, Parliament had intended to implement pre-existing case law on plant and the test for plant in case law was functional.

146.     Ms Nathan’s position was that the explanatory notes to, and the Parliamentary debate of, the CAA showed that Parliament’s purpose was to exclude buildings, structures and alterations to land from capital allowances, other than those which decided case law had already decided were entitled to capital allowances.  It was intended as a descriptive and not functional test as it was not replicating the case-law test for plant but qualifying it. 

147.     I consider that HMRC are right to say that the explanatory notes and, in particular, Hansard indicate that Parliament’s reason for introduction of the exclusions and exemptions to the exclusions was to prevent buildings/structures with long life getting relief intended for items with shorter life and that their intention was to do so by modifying the effect of the case law on plant  So I take from this that Parliament was more concerned with the structure-like nature of the item in question than with its function.  Mr Cordara considered that Hansard could not be referred to as no ambiguity was shown.  But it seems to me that the explanatory notes could be referred to, and in so far as there was ambiguity in whether Lists B and C were intended as descriptive or functional tests, Hansard made it clear that a functional test was not intended. 


HMRC’s case was also that Lists B/C should not be extended by analogy, unless the legislation specifically provided for this by using expressions such as  ‘such as’  or ‘any other structures’.  She referred to the Court of Appeal decision in Henriksen v Grafton Hotel Ltd [1942] 1 KB 82 where the Court said:

…it frequently happens in Income Tax cases that the same result in a business sense can be secured by two different legal transactions, one of which may attract tax and the other not.  This is no justification for saying that a taxpayer who has adopted the method which attracts tax is to be treated as though he had chosen the method which does not, or vice versa.’

Per Lord Greene.

However, that is too general to be helpful in this context.  What I do accept is that List B and C were not intended by Parliament to be a test of function because they were intended as qualifications on what case law had found to be plant, which was a functional test.  Therefore, Parliament did not intend analogies based on function other than where this was made clear. 

List C was definitive

148.     HMRC’s last point was that List C was definitive and it was not possible to say that the gas cavities should be treated as within List C simply because they might be analogous to items in List C, such as storage tanks.

149.     The appellant agreed with this in principle but pointed out that, in its view, the always speaking doctrine would apply to new technology so that the Tribunal had to decide whether the new technology shared the same inherent characteristics as items within the Act of  Parliament.  HMRC did not accept that the always speaking doctrine had any application as gas cavity technology dated back to the 1960s and predated the CAA 1990. 

150.     I agree that List C is definitive and the always speaking doctrine is not applicable as the expert evidence was clear that the technology pre-dated the legislation.  The question is whether gas cavities were one of the items (such as a storage tank) described in List C.

151.     I will consider the application of the exclusions to gas cavities but first mention HMRC’s application to amend their statement of case.

Application to amend statement of case

152.     Shortly before the hearing a dispute arose between the parties because Ms Nathan wanted to include in the bundle authorities on the meaning of structure.   S22(1) excluded certain structures and alterations to land; however, List B paragraph 7(b) contained an exemption for structures ‘in use for the purposes of an undertaking for the extraction, production, processing or distribution of gas’.  The appellants’ case was that they fell into this exemption.

153.     Mr Cordara said the authorities Ms Nathan wished to rely on were irrelevant because HMRC had not pleaded that the cavities were not structures.  HMRC’s position was that it was their case that the cavities were not structures; at the start of the hearing Ms Nathan said that this was already a part of HMRC’s pleaded case but if HMRC did need to amend their statement of case to include this argument, then she was applying to do so.

154.     My ruling in the hearing was that I did not think that HMRC had pleaded a case on this.  Ms Nathan said it was implicit in HMRC’s statement of case but I found that all the statement said on this was that the cavities were not in List B Item 7 because ‘a gas storage facility does not fit comfortably within the description of a permitted structure given that it neither extracts, produces, processes nor distributes gas.  At most, it stores gas prior to its distribution’. I considered that any reasonable reader would only take from this that HMRC did not accept that the cavities were not ‘in use for the purposes of an undertaking for the extraction, production, processing or distribution of gas’:  there was nothing to alert the appellant to HMRC’s case that the cavities were not in List B item 7(b) because they were not structures.

155.     The next question was whether I should permit HMRC to amend their case on this.  But their case was not presaged in their statement nor in their skeleton.  Even at the hearing, the appellant did not know why HMRC considered that the cavities were not structures.  I considered the appellants taken by surprise.

156.     I also considered that they were disadvantaged by the lack of warning.  Ms Nathan had indicated that she intended to ask Mr Bebbington questions relevant to HMRC’s case that the cavities were not structures so even HMRC saw it as something on which evidence could be given (even though Ms Nathan also described it as a pure legal point).  The appellants, however, had been given no chance to consider whether to call evidence on the matter as they had had no forewarning of HMRC’s case on this.

157.     So, in the hearing, I refused HMRC permission to amend their statement of case. It was a very late amendment.  The appellants were prejudiced by not knowing HMRC’s case on this in advance as they had not been able to prepare their legal or factual case to meet this challenge.  It was unfortunate that the appeal would have to continue on the assumption that the cavities were structures and so this aspect of the legal position would be given no consideration, but that was necessary in order to ensure the proceedings were fair to both parties.

158.     So the List B question for the Tribunal would only be whether the cavities in use for the purposes of an undertaking for the extraction, production, processing or distribution of gas; the Tribunal would assume the cavities were structures. And I move on to consider that question.

An undertaking for the extraction, production, processing or distribution of gas

159.     As stated above, HMRC’s position was that the expenditure on the gas cavities was excluded by s 22 because the gas cavities were structures within List B. All structures outside items 1-6 were excluded unless they fell within paragraph 7 (a), (b) or (c).  It was agreed (a) and (c) were inapplicable.  Item 7(b) was, as I have said:

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

HMRC said the cavities were not such structures.  Their case on this was two fold:

(1)     The cavities (said HMRC) were not in use for the purposes of an undertaking;

(2)     The cavities (said HMRC) were not used in the distribution or processing of gas (and it was accepted, and indeed obvious, that they were not used for the extraction or production of gas).

Meaning of undertaking

160.     I think both parties were agreed that the word ‘undertaking’ had two possible meanings:

(1)     One meaning referred to an entity (typically a company, partnership or sole trader);

(2)     The other meaning referred to the doing of an action or task.

161.     What they did not agree on was which meaning was intended in the context of List B.

162.     HMRC’s case was that it was the second meaning which was intended.  HMRC point out that if Parliament had used the word ‘undertaking’ to mean ‘entity’, it would have been more grammatical to say:

…a structure in use for the purposes of an undertaking whose business includes the extraction production processing or distribution of gas….

Further, HMRC pointed out that List B paragraph 7(c) used a similar sentence construction and referred to ‘a trade which consists in the provision of telecommunication…’; their inference was that in both (b) and (c) Parliament was describing what was done, rather than who was doing it.  If HMRC was right, it would mean the question posed was the use of the cavities themselves rather than what was the business of the entity using the cavities.

163.     The appellant thought that the first meaning was intended.  It thought ‘undertaking’ referred to the nature of the business undertaken by the taxpayer.  It relied on an Australian case of Top of the Cross Pty Ltd (1980) 50 FLR 19 which had held that although the meaning of ‘undertaking’ would come from the context in which it was used, its typical meaning was business or enterprise.

164.     The appellant also said that the current legislation and legislative history showed the word took the first meaning.   S 274(1)(b) CAA, and its predecessor s 18(1) CAA 1990, referred to undertakings and, said Mr Cordora, clearly intended to refer to undertakings as entities.  Ms Nathan did not agree and pointed out that both these sections referred to a ‘tunnel undertaking’, a ‘bridge undertaking’ and a ‘dock undertaking’ which she inferred was using undertaking in the ‘action’ rather than ‘entity’ sense.

165.     I would point out that the heading to Table B in s274 was:

Undertakings which are ‘qualifying trades’ if carried on by way of trade

which seems to me to indicate, similarly with the phrases like ‘bridge undertaking’ that Parliament intended undertaking in s 274 to refer to an action rather than an entity.  It seems right to suppose, where there are no contrary indications, that a word has the same meaning wherever used in the same statute, and therefore, also taking into account the point on grammar made by Ms Nathan, I consider ‘undertaking’ in the context of List B was intended to take the second meaning. 

 Cavities used for the purposes of an undertaking for the …distribution/processing of gas?

166.     The next logical question was whether the cavities were used for the purposes of an undertaking for the (in the sense of the action or task of) processing or distribution of gas?

167.     HMRC’s case was that the cavities were used to store gas.  They were not, said HMRC, therefore for the purpose of an undertaking for the distribution or processing of gas.  HMRC went further and said that the NTS was not itself undertaking the distribution or processing of gas.  The NTS was (relying on Mr Aron’s and Mr Bebbington’s evidence) part of gas transmission (high pressure gas moved from processing to distribution pipes) and not a part of the gas distribution network (lower pressure pipes which took gas from the NTS and delivered it to users).

168.     Ms Nathan also said that at the time the legislation was drafted, gas storage, even gas storage in cavities was a known technology (and I accept that that is so on the evidence), and so it had to be presumed Parliament’s failure to include ‘storage’ in the list was a deliberate choice.

169.     The appellants disagreed and thought it acte clair that the assets were used in the distribution and/or processing of gas.  The appellants processed the gas before returning it to the NTS; by returning the gas to the NTS, the appellants were distributing it.  More generally, the appellants saw their ability to store gas as being part of the national gas distribution network.  The gas could be quickly put back into the NTS if required.


170.     I will deal with the question of processing first.  The appellants did process the gas.  They had to process the gas to restore it to the same condition that it was when taken off the NTS.  Hydrates and condensates which had collected in it during storage had to be removed and its temperature might need adjusting.  It seems to me that, whichever meaning is attributed to ‘undertaking’, the cavities were not used for the purposes of an undertaking for the processing of gas. 

171.     If (as I consider right), ‘undertaking’ is understood to mean a task or action, the cavities were not used for the task or action of processing the gas.  The processing was not done by or in the cavities but in the processing plant above ground.  Nor did the cavities store the gas so that it could be processed; it was the other way around.  The processing was an incident of the storage.

172.     And if ‘undertaking’ is understood to mean an entity, then the cavities were not used by an entity for the processing of gas.  The appellants’ purpose was not to process gas.  The appellant’s  purpose was to store gas.  They used the cavities for this purpose.  The appellants did process gas, but only as necessary incident of their storage of the gas.  It was not their purpose. 


173.     So were the cavities used for the purposes of an undertaking for the distribution of gas? As I said above, the question is what Parliament can be supposed to have intended from the context in which the word to be interpreted was used.  While I accept that in the industry, as both Mr Aron and Mr Bebbington said, a distinction is drawn between transmission and distribution, and the NTS is the transmission system and not seen as a part of the distribution system, I do not think that Parliament used the word ‘distribution’ in that technical sense.  The context of the legislation as a whole is legislation which does not use words in their industry-specific technical sense.  It makes little sense to suggest that Parliament intended to use terminology specific to the gas industry in a section which covers all industry as a whole, and I do not think that they did.

174.     I consider that the NTS is a part of the ‘distribution’ of gas as encompassed by List B Item 7(b).  But are the cavities a part of the distribution system of gas?  The cavities themselves do not distribute gas:  on the contrary, they store it.  Nor are they used to support the distribution system:  they are used to fast-cycle gas in order to make money from price volatility.

175.     If (as I consider right), ‘undertaking’ is understood to mean a task or action, the cavities were not used for the task or action of distributing the gas.  The cavities did not distribute gas but store it.

176.     If undertaking should be understood as meaning an entity, however, then it is a more difficult question to answer whether the cavities were used by an entity for the distribution of gas.  On one level, it could be said that the appellants’ purpose was not to distribute gas but to make profits by storing it. However,  CCS could only make money from storing the gas by selling the gas back onto to the NTS for onward distribution. EDF made its money by allowing fast cycle storage of gas belonging to its customer so the ability to sell the gas back onto the NTS so that it could be distributed was also crucial to EDF’s business.

177.     I do not have to resolve this particular dispute because I have already said that ‘undertaking’  in this context means task or action and not entity, but if I had to resolve it, I would have said that the appellants were entities for gas storage and not for gas distribution.  While they both existed to enable stored gas to be sold back onto the NTS, they did not themselves distribute gas by transmitting it to users or other transmitters.  They simply sold the gas they stored.

178.     My conclusion is therefore that the cavities were on List B as ‘any structure’ and were thereby excluded from benefit of capital allowances.  They were not saved by falling into Item 7(b) as they were not structures in use for the purposes of an undertaking for the extraction, production, processing or distribution of gas. 

179.     It is therefore not relevant to consider whether the cavities were also excluded from the benefit of capital allowances by considering whether s 22(1)(b) applied, but for the sake of completeness I will do so.

 Works involving the alteration of land

180.     Two issues arose under this heading.  Firstly, if the gas cavities were not excluded under s 22(1)(a), could they be excluded under s 22(1)(b)?  As a reminder, s 22(1) contained two exclusions:

(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.

If (a) and (b) were mutually exclusive, because HMRC had in effect conceded that the cavities were structures, and List B in effect applied to ‘any structure’, the cavities could not fall into s 22(1)(b) as works involving the alteration of land.

181.     The second issue was whether the works undertaken (leaching and de-brining) actually involved the alteration of land.  I will take these two issues in turn.

Are s 22(1)(a) and (b) mutually exclusive?

182.     Mr Codara’s position was that s 22(1)(a) and (b) had to be mutually exclusive because virtually everything in List B involved the alteration of land and so List B would be pointless unless (a) and (b) were mutually exclusive.  Judge Poole had recently come to a similar conclusion in SSE Generation at [39-40]:

One further point of significant argument between the parties was around the approach to interpreting section 22(1)(b) CAA.  Mr Brennan argued that anything which involved alteration of land should fall within the provision, and it did not matter whether the relevant item was a structure or asset also falling within List B.  Mr Peacock argued that the effect of this approach would be to make List B entirely redundant, as the provision of all of the structures and assets referred to in it would necessarily involve alterations of land in some way.  This could not have been the draftsman’s intention, and the correct approach to make sense of section 22(1)(b) CAA must be to interpret it as applying to civil engineering works involving the alteration of land with a result which does not fall into List B.  Support for this approach was given by the threefold distinction between “structures”, “assets” and “works” explicitly stated in the heading of section 22 and implicit in the phraseology of sections 22(1)(a) and (b). Were one to do otherwise, then there would also be an obvious irreconcilable conflict between section 22(1)(b) CAA and the exclusions to the general “catch all” Item 7 of List B – taking for example a gas extraction structure involving alterations to land, which would be disqualified from allowances under Item 7 but for the saving in Item 7(b).  It was inherent in Mr Brennan’s argument that it would nonetheless be disqualified from allowances by reason of section 22(1)(b) CAA, simply because it involved alterations to land; that could not be right.  Mr Brennan did not shy away from the point.  He submitted that the boring of a tunnel, for example, would fall within both section 22(1)(b) as “works involving the alteration of land” and List B Item 1.  If the tunnel were then lined with bricks, the expenditure incurred in doing so would only fall within List B Item 1 and not section 22(1)(b) CAA. 

40.              I prefer Mr Peacock’s approach in principle.  I consider sections 22(1)(a) and (b) CAA to be alternatives, not largely overlapping, and that the “works” referred to in section 22(1)(b) CAA 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.

183.     HMRC did not consider SSE Generation rightly decided and had appealed it.  Ms Nathan referred me to the decision in Maco Door and Window Hardware (UK) Ltd [2008] UKHL 54 where, in respect of a different part of the capital allowances legislation, the House of Lords concluded that certain categories could not be mutually exclusively precisely because one was a subset of the other. 

184.     Mr Cordara’s response was that the provisions in question in Maco Door granted allowances; here s 22 excluded assets from the benefit of allowances so, he said, it would make little sense for one to be the sub-set of the other.

185.     I agree with Judge Poole’s analysis.  Parliament was careful in Item 7(a)-(c) to grant exemptions from the general exclusion for structures and assets in s 22(1)(b).  That section of the legislation would be rendered pointless if those exempt assets fell foul of s 22(1)(b).  As that cannot have been intended, it must have been intended that structures or assets within s 22(1)(a) were not works within s 22(1)(b).

Were the works an alteration of land?

186.     I do not therefore need to consider whether the leaching and de-brining were ‘works involving the alteration of land’ but for completeness will do so as the point was argued.

187.     The question of whether the leaching was works involving the alteration of land is easily answered as the appellant concedes that it was; I think it a concession well made as the introduction of water into halite dissolved it and created an underground cavity where there was previously none.  The land was clearly altered.  I note in passing that the creation of underground cavities in SSE Generation  was also considered to be works of alteration to land (albeit ‘land’ in that context just carried the unmodified Interpretation Act meaning):

[108] ‘In my view, the caverns clearly fell outside any of the Items in List B,,,but quite clearly amounted to “works involving the alteration of land”

188.     But did de-brining alter the land?  HMRC said it did because what was an underground cavern containing salt water became an underground cavern containing gas. Removing the brine improved the land by making the cavity fit for gas storage.   The appellant said de-brining did not alter the land:  the cavity itself remained unaltered whether it contained brine or gas.

189.     Mr Cordara referred to the definition of land applicable to s 22.  By s 22(3)(b) it was stated that:

‘land’ does not include buildings or other structures, but otherwise has the meaning given in Sch 1 to the Interpretation Act 1978

His interpretation of this was that ‘land’ was therefore the soil and rock; it did not include brine in a cavity and so there was no alteration to land by removing brine from a cavity. 

190.     I am not able to agree with the appellant on this.  The evidence was that the cavity was unlikely to be stable if left empty; the brine or the cushion gas with which it was replaced had a structural purpose and, I consider, should be seen as part of the land.  So when the brine was exchanged for gas there was an alteration to the land.  In any event, works, say, to drain land would be considered works altering land so it seems that de-brining a cavity is also work altering land.

191.     So, if I had not concluded that s 22(1)(a) and (b) were mutually exclusive, I would have held the costs of leaching and de-brining were excluded from capital allowances under s 22(1)(b) as works involving the alteration of land.  But I do consider them mutually exclusive; nevertheless, although it is conceded by both parties that the cavities were structures so that s 22(1)(a) applied, I have held that the cavities do not fall into paragraph 7(b) as structures in use for the purposes of an undertaking for the extraction, production, processing or distribution of gas and so they would be excluded from the benefit of capital allowances under s 22(1)(a).

192.     Having said that, as I have explained above, whether the cavities were excluded under s 22(1)(a) and/or s 22(1)(b) (as I have found that they were) does not matter if they fell within the exemptions set out in List C.  So I will consider each of the three potentially applicable exemptions, although, as I have said, this section on the exclusions and exemptions from exclusions is all hypothetical as I have concluded that the cavities were not plant.

List C – item 4 - Storage equipment?

193.     The first relevant item of List C was no. 4 which applied to:

4.  Manufacturing or processing equipment; storage equipment (including cold rooms); display equipment; and counters, checkouts and similar equipment.

While I have set out no 4 in full, in reality the appellant’s case was that the gas cavities were ‘storage equipment’.

194.     While HMRC accepted (rightly) that the gas cavities were for storage, they did not accept that they were ‘equipment’.   HMRC relied on the definition in the Shorter English dictionary that

Articles used or required for a particular purpose; apparatus

HMRC’s position was that underground cavities could not be described as articles or apparatus, however much they might have a purpose.

195.     The appellant’s position was that the cavities were equipment because they were, or served as, pressure vessels for storage of gas.  Moreover, if a cold room is ‘equipment’ it naturally followed that another storage structure, such as a gas cavity, was also equipment.  They were both permanent, fixed man-made spaces for storage under particular conditions.

196.     I’m going to consider what Parliament meant by the use of ‘equipment’ in this context by considering the normal meaning of ‘equipment’ and then whether a different and wider meaning was intended because it was said to include cold rooms.

197.     With the exception of the reference to cold room, everything else in item 4 is some sort of ‘equipment’ in the normal meaning of the word even though some of them might be well be fixtures when used (eg manufacturing equipment and counters).  Were it not for the reference to cold rooms, I would consider that Parliament intended ‘equipment’ to be used with its normal meaning. 

198.     It follows, therefore, that although Mr Aron’s evidence was that industry would not consider a cavity to be equipment, I do not consider that matters as the question is its normal and not technical meaning.  I am not clear whether Mr Aron’s view that equipment had to have moving parts was a personal view or a view which reflected that of industry, but I do not accept that it is relevant.  The normal meaning of equipment would not require moving parts, only that it is an article with a purpose.

199.     However, I agree with HMRC that the word ‘equipment’ would not in normal usage encompass an underground cavity, even one that was man-made for a particular purpose.  The word ‘equipment’, as well as the words ‘article’ and ‘apparatus’ used to define it,  imply a thing that is not a part of the land (although it might include something that has been fixed to the land in order to use it).  Equipment may not be portable but it is not a part of the landscape.  The cavity, however, is formed in rock. It is part of the land and not fixed to it.

200.     But should ‘storage equipment’ be understood to have a wider than normal meaning as it is stated to include cold rooms, which, if fixed, are a part of land?  I would not consider a cold room to be an article. I would assume it was a purpose built structure or part of a structure with design features which enable (normally only when powered) a cold temperature to be maintained.  So should other structures with a storage function be considered storage equipment?

201.     It seems to me that drafter understood that ‘storage equipment’ would not include cold rooms, and so specifically included it as Parliament wished cold rooms to have the benefit of capital allowances (perhaps to reflect the Union  Storage case).  I do not think there was an  intention to widen the meaning of ‘equipment’ generally as the drafter said ‘including’ cold rooms and not ‘such as’ cold rooms.  I have already said I do not think Parliament meant the exemptions to be widened by analogies with function.

202.     My conclusion is that item 4 is fairly narrowly drawn and would not ordinarily extend to structures with a purpose although it would appear to extend to fixed apparatus.  While it was clearly intended to extend to cold rooms, it was not intended to extend to store rooms, or it would have said more than simply ‘including cold rooms’. So Item 4 does not cover the cavities as they are not equipment.  They are part of the sub-surface rock.

203.     But even if I am wrong and it is possible to extend List C by analogy and so other structures with a function like a cold room should benefit from the exemption, I do not think that the cavity functioned as a cold room.  A cold room is designed to operate on the thing stored by reducing its temperature.  Cavities were not designed to operate on the gas stored other than in the sense that they were designed to retain it.  While they might have an effect on temperature, it was not their purpose nor a significant effect.  And while they were designed to store gas at greater than atmospheric pressure, that did not change the gas in any way but merely allowed a large quantity to be stored; moreover (unlike a cold room) the cavities did not require any parts or power to do so.  It was just the natural effect of a large hole in salt rock.  I do not think the cavities are analogous to cold rooms.  My conclusion is that the cavities do not fall within item 4 of  List C.

List C - Silos or storage tanks?

204.     It makes sense to then look at the last potentially relevant item in List C, no. 28, as that was:

28. the provision of –

(a) silos provided for temporary storage, or

(b) storage tanks

205.     The reference in 28(a) to silos provided for temporary storage appears to be a reference to the decision in Schofield and to show Parliament intended that the new exclusions would not affect existing caselaw.  As I understood it, the appellants did not suggest that their cavities were silos and so 28(a) is not really relevant, other than perhaps to the appellants’ case that List C should be extended by analogy. 

206.     I agree that the cavities were not silos as the dictionary definition of silos shows that its normal meaning is:

 A pit or underground chamber used for the storage of grain, roots, etc.

and it is not apt to describe cavities for storage of gas.

207.     So far as the question of analogies are concerned, I agree with HMRC that List C is meant to be descriptive and not to be extended by analogies on function.  In other words, something which does not meet the description but which does carry out a similar function was not intended by Parliament to be in List C as the purpose of the exemptions to the exclusions was to preserve the outcome of decided cases on plant but to reverse the perceived trend in cases of long term structures receiving capital allowances. 

208.     It was that appellants’ case that the cavities were storage tanks.  I agree that the cavities were for storage:  I do not think that that was in dispute but it is clear on the evidence in any event.  But were they tanks? I was referred to the OED definition for tank which was:

Artificial receptacle, usually rectangular or cylindrical and often of plate- iron used for storing water, oil or other liquids in large quantities.

209.     The appellants were not concerned that the cavities were not rectangular or cylindrical:  the definition only said tanks were ‘usually’ such shape.  Further, they did not consider that it was relevant that they were not made of plate iron as the definition only said tanks were ‘often’ so made.  I agree and HMRC did not suggest otherwise. 

210.     The cavities were also clearly artificial.  They were man-made by a process of leaching. I also do not think their size relevant:  tanks could be very large (see BP Refinery v Walker).

211.     HMRC’s case that they were not tanks rested on their view that tanks had to be made or manufactured out of something and the cavities were not really made of anything:  they were a hole in the ground, an absence of rock.  They were not even lined.  And a further issue was that the above OED reference clearly regarded a tank as used for storing liquids.  Gas is a fluid but not a liquid. 

212.     However, the Shorter OED definition of ‘tank’ to which I was also referred was:

A (large) receptacle or storage chamber for liquid or gas

Nevertheless, the only reference to a ‘gas tank’ in the OED appeared to be from an American source and to refer to a petrol tank, and therefore to be for storing liquid.  I also bear in mind that Mr Aron said that he would not regards gasometers as tanks and indeed a book about them referred to them as gas holders.  My conclusion was that it was not a common use of the word ‘tank’ to refer to it as a container for the storage of gas rather than liquid.

213.     Having said that, I was also aware that prior to use as gas cavities, the cavities had held liquid, as they held brine, albeit not for storage.  Storage is not the only purpose of a tank:  for instance, a fish tank is a tank and it purpose is to retain the water necessary to the life of the fish.  So would the cavities properly be described as tanks when they held the brine?  Because, if so, it would be odd if they ceased to be tanks when the brine was exchanged for gas.

214.     But I did not think it a natural use of the word ‘tank’ to describe a cavity in rock which was full of brine, particularly when the purpose of holding the brine in the cavity was not for retaining the liquid but to maintain the integrity of the cavity. 

215.     So when gas was exchanged for brine, when the purpose of the cavities became one of retaining a gas, did that mean that they were or became tanks? While it is possible that the word ‘tank’ could be used to describe a container holding a gas, it is normally only used for liquids; and while the etymology of ‘tank’ is concerned with cisterns, which can be underground and cut from rock, cisterns are only for water and the modern meaning of tank does not really encompass a cistern, and even if it did, it would not be something for holding gas.  I came to the conclusion that it would not be a natural use of the word ‘tank’ to describe an underground chamber formed out of the rock and used to store gas.  Moreover, the legislation post-dated the technology so there was no need to refer to the always speaking doctrine.  

216.     My conclusion is a gas cavity was not a ‘storage tank’ in the meaning of the legislation and therefore the cavities did not fall into List C item 28.

List C- Item 22 - Alteration of land to install plant or machinery?

217.     The other relevant item in List C was no. 22:

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

218.     Both parties were agreed that leaching was an alteration of land; I have found that de-brining was also an alteration to the land.

219.     So the question is whether the leaching and de-brining were for the purpose only of installing plant.  There was no suggestion they were for the purpose of installing machinery.

220.     The appellant’s position was that the cavities were plant; so the work to create the cavities was work to install them.  I have ruled that the cavities were not plant, but putting that on one side and assuming that they were, HMRC say that it is a misuse of language to say that the land was altered for the purposes of installing plant.  The alterations did not install the cavities: the leaching created the cavities and the de-brining exchanged their contents for other contents.

221.     The appellant relied on SSE Generation where the judge said:

[78]….the answer to the question revolves around the meaning of the word “installing” and in particular whether it extends to include installation by the creation in situ of the asset in question, in addition to installation by putting in place something which previously existed, albeit perhaps only in component form (as in the case of installation of a pipeline.

[79] The OED relevantly defines ‘install’ as ‘to place (an apparatus, a system of ventilation, lighting hearing, or the like) in position for service or use’.  This does not take matters much further.  I consider the matter finely balanced, but standing back and looking at the matter realistically, the end result of the appellant’s operations was to create in the appropriate place an item of plant (the aqueduct) which was an important element of the overall Scheme where previously they had only been solid rock. Looked at in that way, I consider that alteration of land involved in the creation of the aqueduct to have been carried out for the purpose only of installing the aqueduct.

222.     However, I am unable to agree with Judge’s conclusion.  ‘Install’ carries the implication that something pre-existing is put in position.  Creating a space where previously there was none does not install the space.  It creates it but it does not install it.   And contrary to what the appellant says, I do not consider that either the dry dock or swimming pool were installed (respectively) in Barclay Curle or Beach.  In each case, the item of plant was created in situ.  If the cavity was an item of plant, the leaching resulted in its creation; it did not install it in the natural meaning of the word.  The de-brining created a cavity suitable for gas storage but it did not install such a cavity.

223.     Ms Nathan referred to the Court of Appeal decision in Engineering Industry Training Board v Foster Wheeler john Brown Boilers Ltd where it was said:

[‘installation’] conveys putting in place something already made so that it can be used.  There may be an element of assembly required; but basically a thing installed is ready to work when it is put in its place, and, if necessary, connected up. ….Installation seems to me to be very far from the reality of the facts, ….It think it is truly construction.

In that case, lots of different components of a very large boiler were pre-manufactured and brought to site and the boiler was assembled in situ.  It was found not to be installed.  This fortifies my view that creating the cavities is not the same as installing them.  Something could be installed only after it was created.

224.     Judge Poole’s analysis in SSE Generation appeared to be one which relied on analogy in the sense that creating an aqueduct was of similar effect to installing an aqueduct, which may or may not be correct, but I have already said that I think HMRC are right to say that List C was not meant to be extended by analogy.

225.     In conclusion, even if I had found the cavities to be plant, I would be unable to accept that they were installed.

226.     I suggested in the hearing that, as it was accepted by HMRC that cushion gas was plant, it might be possible to say that the cavities were created for the purpose of installing the cushion gas.  However, I accept that Ms Nathan was right to say that my point did not assist the appellants because they would have to show that the cavities were created

….‘for the purposes only of installing plant …’

and it is clear that the cavities were not created only for ‘installing’ the cushion gas (even presupposing it is right to say that the cushion gas was installed).  And that is because the cavities were actually created to store working gas (stock in trade).  While they had to be first filled with cushion gas to make them stable, the reason they were created was to allow storage of working gas.  The land was not altered by the leaching and de-brining for the purpose only of installing plant.

227.     My conclusion is that the works to create the cavities do not fall within item 22 of List C.

Overall conclusion

228.     My overall conclusion is that the appeal fails for two distinct reasons.  Firstly, I have found for the reasons given that the cavities were not plant; secondly, I have found, for the reasons given, that even if the cavities were plant, they were excluded from allowances by s 22(1)(a) and were not saved by List C.


Right to apply for permission to appeal

229.     This document contains full findings of fact and reasons for the decision.  Any party dissatisfied with this decision has a right to apply for permission to appeal against it pursuant to Rule 39 of the Tribunal Procedure (First-tier Tribunal) (Tax Chamber) Rules 2009.  The application must be received by this Tribunal not later than 56 days after this decision is sent to that party.  The parties are referred to “Guidance to accompany a Decision from the First-tier Tribunal (Tax Chamber)” which accompanies and forms part of this decision notice.



Barbara Mosedale



Release date: 1 August 2019


BAILII: Copyright Policy | Disclaimers | Privacy Policy | Feedback | Donate to BAILII