The Judicial Committee of the Privy Council Decisions

Auckland Gas Co. Ltd. Appellant

v.

The Commissioner of Inland Revenue Respondent

FROM

COMMITTEE OF THE PRIVY COUNCIL,

Delivered the 14th June 2000

------------------

Lord Nicholls of Birkenhead

[Delivered by Lord Nicholls of Birkenhead]

------------------

1. This appeal concerns the deductibility of expenditure incurred by the appellant taxpayer on its low pressure gas distribution pipeline system for its five income years ending June 1988-92. The taxpayer, Auckland Gas Co. Ltd., claimed that in accordance with the provisions of the Income Tax Act 1976 it was entitled to deduct this expenditure when calculating its assessable income. The Commissioner of Inland Revenue disallowed the claim. The trial judge, Williams J., disagreed with the Commissioner and upheld the claim of Auckland Gas (1997) 18 N.Z.T.C. 13408. The Court of Appeal, comprising Richardson P., Gault, Henry, Blanchard and Tipping J.J., [1999] 2 N.Z.L.R. 418 took the contrary view and allowed the Commissioner's appeal.

 

2. The decisions of both courts below are reported. It is therefore unnecessary for their Lordships to set out the background facts in detail. These are stated very fully in the judgment of Williams J. The following summary is sufficient for the purposes of this appeal.

 

Auckland Gas and its distribution system

3. For many years Auckland Gas had a mainly low pressure network of underground cast iron pipes, known as mains, and smaller steel spurs, known as services, running from the mains to individual premises. Cast iron mains had a life of a century or more, but their use was by no means trouble-free. Two perennial major problems were leakage of gas and the entry of water. This occurred at joints in the mains, at fractures in the mains and at corroded parts of the services. Production of gas-tight joints was always a difficult problem. Cast iron pipes were manufactured in comparatively short lengths, up to twelve or eighteen feet, and there were many thousands of joints in a network the size of Auckland Gas. Further, fractured mains were a weekly occurrence for Auckland Gas. Cast iron pipes fractured by being left unsupported when the surrounding ground subsided or by being damaged by vibration from passing traffic, nowadays of a weight and frequency not contemplated when the mains were laid. Corrosion was a problem with the steel services, especially those manufactured before coated steel was used.

 

4. Over the period from 1900 to 1939 the percentage of gas entering the system but then lost ("unaccounted for gas", or "UFG") varied between 6% and 17% annually. The figures deteriorated thereafter. Insufficient resources were devoted to maintenance, and in the 1950s and 1960s the yearly UFG figure was usually over 20%. The worst year was 1971 when no less than 38.5% of the gas entering the system was lost.

 

5. The UFG problem was exacerbated in the 1970s by the introduction of natural gas. Lengths of cast iron pipe were mostly joined by bell and spigot joints, filled with hemp kept in place by molten lead. Until about 1970 the gas supplied by Auckland Gas was coal gas. Coal gas is a wet gas, and the moisture and tar residues from the gas provided a sealant for the hemp joints. Natural gas, which came on-stream from Taranaki in 1970, was a dry and clean gas. When natural gas passed through the cast iron system the joints dried rapidly, causing a lack of gas-tightness and more leaks.

 

6. Until the early 1980s most new mains and services laid to distribute natural gas directly to consumers in Auckland were made of coated steel rather than cast iron. Then Auckland Gas began to use polyethylene pipes for all new mains and services. Polyethylene piping is tough and reasonably flexible. It is resistant to abrasion, to pressure stress and to corrosion. It does not require a protective casing. It is relatively easy to work, and is supplied in long lengths of 150 metres or so. It can be joined by the application of heat, and the joints are as strong as the pipe itself. While not impervious to damage, it is largely unaffected by ground movement arising from vibration, subsidence and the like and so is much less affected by traffic, earthquakes and so on than cast iron. It has a life span of fifty years or more. When laid as new mains, polyethylene piping was placed somewhat further underground than the old style mains and was surrounded by soft sand or a thin metal casing which provided support and some strengthening.

 

7. By 1986 Auckland Gas’s existing low pressure system was in a poor state of repair. The system had become very expensive to maintain by "find and fix" repairs at each individual joint and at each corroded or fractured piece of pipe. The system was also unreliable, because the ingress of water led to breakdowns and temporary cessation of supply to customers. This hindered the company's ability to market its product.

 

8. After attempts over many years to develop other techniques, Auckland Gas adopted the technique of inserting a polyethylene pipe into the existing cast iron mains and steel services. This reduced the volumetric capacity of the system because of the smaller bore, but gas could be transmitted in the polyethylene pipes at a much higher pressure than through the old cast iron mains. Insertion was achieved by breaking open the mains at intervals, removing projections likely to damage the polyethylene pipe, and pushing the pipe down the mains. The old mains were also broken open to allow the polyethylene services to customers' premises to be connected to, and sealed with, the new polyethylene main. Thus, the old leak-prone joints were outside the pipe containing the gas, and they became irrelevant. The inserted pipe also cut off old mains and service pipes which were no longer being used to supply customers but were a dangerous source of leakage. In the first three years of the insertion programme this led to a reduction in overall mains length of about 10%. The only continuing function of the old mains was to act as a "carrier" or conduit for the polyethylene piping and provide it with some support equivalent to the sand or thin metal in a new reticulation. If the old main was less than 40 cm. below the surface, insertion was not undertaken. This insertion technique solved the UFG problem much more cheaply than the traditional "find and fix" digging and repair.

 

The issue

9. The sole issue before the Board is whether the expenditure on the polyethylene pipe insertion programme for the years in question was a revenue item. If it was, the expenditure was within section 104 of the Income Tax Act 1976 and not excluded as capital expenditure by section 106(1)(a). The parties were agreed that the expenditure would be a revenue item and deductible under section 104 if, but only if, the work constituted repair as distinct from replacement. Before the courts below there was a further issue concerning the application of the second proviso to section 108, but that issue was not pursued before the Board.

 

The judgments of the courts below

10. Williams J. accepted the contention of the taxpayer company. He looked at the overall position before and after the polyethylene insertion programme. Beforehand, Auckland Gas had a network capable of delivering enough natural gas to satisfy demand but suffering from an unacceptable level of UFG and requiring extensive and costly maintenance. Afterwards, the company had a network similarly capable of supplying sufficient natural gas at medium pressure to meet demand. It was neither more extensive nor longer lived but it had the major advantages of significantly reduced installation costs, being virtually free of UFG and having much lower repair costs than the existing network. Although, in the judge’s words, the method of repair "differed from what might ordinarily be regarded as repair", in its essence the polyethylene pipe insertion programme was a repair and should be charged to revenue.

 

11. The Court of Appeal disagreed. Blanchard J. gave the principal judgment. He pointed out that the problem confronting Auckland Gas was a maintenance problem. If work of a maintenance character had been done on the joints, and if sections of corroded pipe had been replaced, the cost would have been deductible. Then no new asset would have been created. It would simply have been the old pipeline repaired and restored. But that is not what occurred. In its nature and scale, the polyethylene insertion programme could not realistically be regarded merely as a repair of the network. For a leaky cast iron and steel low pressure system which it was not worthwhile renovating in its existing form, there had been substituted a system constructed of a new and different material, operating at a much higher pressure. The old mains and services were effectively abandoned as a conveyor of gas and relegated to the role of a housing for the new pipeline. The old mains and services no longer performed any function save protecting the polyethylene pipe against vibration and ground movement. That could not fairly be described as a method of fixing the faulty joints and corrosion. Blanchard J. said at page 433 [1999] 2 N.Z.L.R. 418:-

"If a taxpayer faced with costly maintenance bills elects to rebuild the asset in question in a different way which results in a substantial change in its character, the cost cannot be written off for tax purposes as if it were merely an expenditure on maintenance work which was not actually done - even if what was spent may have been less than the cost of the maintenance. Nor is it in point that the asset may give no greater performance and that its potential lifespan may be no greater if in truth a different and substantially improved asset has been created by the expenditure."

 

12. Gault J. at page 436 delivered a concurring judgment, adding some observations on the difficulties which may arise in identifying what is the appropriate asset to be looked at when determining whether expenditure is to be classed as capital or revenue. Had it been necessary, use of the comprehensive description "network" would not have deterred him from examining the component sectors of pipe configurations.

 

Repair and replacement

13. Courts have frequently been faced with having to decide whether an item of expenditure is of a capital or revenue nature. Many of the reported decisions are noted in the judgment of Williams J. Given the parties' agreement on the applicable test in the present case, namely, whether the work constituted repair as distinct from replacement, their Lordships need not survey this much explored field yet again.

 

14. Their Lordships therefore turn directly to the question whether the work constituted repair or replacement. In this context these words are not technical expressions with a special meaning. Rather, they bear an ordinary, everyday meaning. They connote repair, or replacement, of a tangible object. In order to decide whether work constitutes repair or replacement, the first step is to identify the object to which the test of repair or replacement is being applied. Frequently this is a straightforward exercise and the answer is obvious. To take a homely instance, replacement of a worn washer on a household tap is normally regarded as a repair of the tap even though one of its parts has been wholly replaced. The tap has been repaired by the replacement of one of its component parts. Similarly with a car: replacement of a spent battery or a corroded exhaust system will normally be regarded as a repair of the car. The car has been put into working condition again. It often happens that, with improvements in technology, a replacement part is better than the original and will last longer or function better. That does not, of itself, change the character of the larger object or, hence, the appropriate description of the work.

 

15. Some objects do not lend themselves so readily to this exercise in characterisation. The nature of some objects and their component elements is such that replacement of one or more components will not necessarily be regarded simply as a repair of the larger object. This is particularly so if the replaced element differs from the damaged original in such a way as to change the character of the whole. A house is a simple example of this. Demolition and rebuilding of a dangerous flank wall of a house would normally be regarded as repairing the house. The answer might not be so obvious if an entire derelict wing of a large house were demolished and rebuilt, especially if the new construction were substantially different from the original. Questions of degree may arise in such cases.

 

16. Authority on the question of repair or replacement is of limited assistance. The physical objects to which the test of repair has to be applied vary widely. So does the nature of the work done. Judicial dicta applicable to one set of circumstances may be unhelpful or misleading when applied in different circumstances. This is true even of the celebrated observation of Buckley L.J. in Lurcott v. Wakely & Wheeler [1911] 1 KB 905, 924:-

"Repair is restoration by renewal or replacement of subsidiary parts of a whole. Renewal, as distinguished from repair, is reconstruction of the entirety, meaning by the entirety not necessarily the whole but substantially the whole subject-matter under discussion." (Emphasis added.)

 

17. As shown by the above example of the demolition and reconstruction of the derelict wing of a house, sometimes repair may not be the appropriate description of work even though it falls far short of being a replacement of substantially the whole of the relevant subject-matter. The effect of the work on the character of the object is also an important consideration.

 

18. This is explicit, or implicit, in several decided cases. In W. Thomas & Co. Pty. Ltd. v. Commissioner of Taxation of the Commonwealth of Australia (1965) 115 C.L.R. 58, 72, Windeyer J. observed that repair "involves a restoration of a thing to a condition it formerly had without changing its character" (emphasis added). In Highland Railway Co. v. Balderston (Surveyor of Taxes) (1889) 2 T.C. 485 parts of the main railway track were relaid, not after their existing fashion, but with steel rails and heavier chairs. The Court of Session held this substitution was a material alteration and great improvement, and contrasted this with taking away any worn rails and renewing them along the line: that "would not alter the character of the line" (see the Lord President, Lord Inglis, at page 488). The Judicial Committee applied this dictum in Rhodesia Railways Ltd. v. Collector of Income Tax, Bechuanaland Protectorate [1933] AC 368, where the cost of relaying a railway line so as to restore it to its former condition was held to be a legitimate charge against income. Consistently with this, in Mitchell v. B.W. Noble Ltd. [1927] 1 KB 719, 729, Rowlatt J. observed that replacement of a railing which perpetually falls down or needs painting with a brick wall would be capital expenditure. In Federal Commissioner of Taxation v. Western Suburbs Cinemas Ltd. (1952) 86 C.L.R. 102 a dangerous ceiling in a cinema was replaced with a new and better ceiling. Kitto J. regarded the work as different in degree and kind from the type of repairs properly allowed for in the working expenses of a theatre business.

 

19. In the last resort, as with the general question of distinguishing income expenditure from capital expenditure, the solution is not to be found by any rigid test or description: see Lord Pearce in B.P. Australia Ltd. v. Commissioner of Taxation of the Commonwealth of Australia [1966] A.C. 224, 264. The answer depends upon a consideration of all the circumstances. They may not all point in the same direction. Then it may be temptingly easy to say the answer is a question of fact and impression. That would be a mistaken approach. The dominant features which guide to a reasoned conclusion must be capable of being identified. Adapting the words of Dixon J. in Hallstroms Pty. Ltd. v. Federal Commissioner of Taxation (1946) 72 CLR 634, 646, the distinction between repair and replacement is not so unascertainable that it must be placed in the category of an unformulated question of fact.

 

The appropriate description of the work done by Auckland Gas

20. In the present case work was done to a gas distribution system. The courts below treated the entirety of the system as the relevant asset. The taxpayer defended this approach, and their Lordships are content to proceed on this footing. But one must be clear about what is meant in this context by Auckland Gas’s distribution system. That expression is not a reference to a means of distributing gas which can be considered as a functional entity separate from its physical components. In the present context Auckland Gas’s distribution system is a reference to an assemblage of linked pipes whose function was to carry gas from one place to another. If a significant portion of this series of linked pipes is effectively abandoned and replaced wholesale with new pipes, the work may readily go beyond what would normally be regarded as repair of the existing system. This is especially so if the new pipes are made of materials which perform differently from the old ones. The work may be of such a nature and scale as to change the character of the existing system. This is to be contrasted with replacing or making good specific leaking pipes or joints. The latter would be repair, the former would do more than repair what was damaged.

 

21. When the polyethylene insertion programme began in 1987 Auckland Gas’s low pressure mains system, comprising cast iron pipes, represented about 70% of the entire mains system. The oral evidence at the trial put the overall length of the mains network at 1,581 km. and the length of the low pressure portions at about 1,100 km. The residue of the mains system largely comprised a medium pressure steel arterial system. This was constructed in the late 1960s and early 1970s, after the building of a medium pressure main over the Auckland Harbour Bridge in 1959 had enabled the company to expand its network to the outer regions of greater Auckland. In the early 1980s this network was extended to supply compressed natural gas to motor vehicle refuelling stations.

 

22. Over the five years in question on this appeal, 1988-92, polyethylene pipes were inserted altogether in 380 km. of the cast iron mains. This represented about 23% of the total mains network. In addition, polyethylene pipes were inserted in 150 km. or 32%, of the steel services. This meant that, so far as these substantial portions of the distribution system were concerned, gas was no longer distributed through cast iron mains or steel services. It was distributed through polyethylene pipes, which were mostly, but not entirely, housed in the old mains and services. These pipes differed from the old in several respects: they were virtually leak-free; they were much better suited to the passage of dry and clean natural gas; and they carried gas at a higher (medium) pressure. Although this latter difference was not a goal in itself, it was an improvement brought about by the changed composition of the pipes.

 

23. Had these new pipes been laid wholly outside the old mains and services, they could not sensibly have been regarded as a repair of the old mains and services or as carrying out repairs to the existing pipeline system. Had this been done, the new pipes, differing significantly from the old, would have comprised a complete replacement of whole portions ("network sectors") of the existing system.

 

24. The course actually followed, of inserting the new pipes inside the old, was not materially different. In future the gas was carried by the new polyethylene pipes, not the old mains and services. The unusual and complicating feature is that the old mains and services, which remained on site, still discharged a useful function: providing support for the new pipes. But, for present purposes, that feature is neither here nor there. The function for which the old mains and services were laid and used was as gas carriers. That was their function as part of Auckland Gas’s distribution system. After the insertion of the new pipes, they no longer discharged that function. They became redundant as gas carriers. Leaks and holes and corrosion in the old mains and services no longer mattered. Far from restoring the gas distribution system to its original state, the work changed the character of the existing gas distribution system: a significant portion of it had been upgraded. Substantial portions of the cast iron mains and steel services were superseded by polyethylene pipes having the differences and advantages mentioned above.

 

25. A further complicating feature is that, not surprisingly, the insertion programme was spread over many years. By June 1992 the work had not been completed, nor was there any certainty it ever would be. The board of Auckland Gas would be bound to keep such a major item of expenditure under review. In their Lordships' view, this feature does not assist the taxpayer. The speed or slowness with which the work was carried out cannot affect its nature or, hence, its proper characterisation. In any event, as shown by the figures already mentioned, by June 1992 the work completed covered a substantial part of the entire mains network. This work was spread more or less evenly over each of the five relevant years.

 

Contrary arguments

26. At the heart of the appellant's submissions, attractively presented by Mr. Farmer Q.C., was an argument that on these matters the impressions and assessment of the evidence by the trial judge should be preferred, notwithstanding the Court of Appeal might regard the evidence and its significance differently. He prayed in aid the decision of the Board in Rangatira Ltd. v. Commissioner of Inland Revenue [1997] 1 N.Z.L.R. 129. Their Lordships are unable to accept this line of argument. It is clear that the Court of Appeal was mindful of its limited role as an appellate court. Blanchard J. noted that the borderline between a repair and replacement of a capital nature may be difficult to determine and that an appellate court should not overturn the choice of the trial judge unless it concluded that his or her assessment cannot be supported by reference to the evidence as a whole. The Court of Appeal held, and their Lordships agree, that Williams J. reached a conclusion which did not reflect the reality of the work done. In particular, his comparison of the functional position before and after was made at a level of abstraction which paid insufficient regard to the nature and extent of the operation carried out by Auckland Gas. A maintenance problem such as existed here may be capable of being solved in more than one way. It may be solved by work which would be regarded as a repair of the existing structure. Or it may be solved by scrapping all or much of the existing structure and providing a new one. In overall functional terms the result may be much the same in the two cases, but that is not by itself a reliable guide. If the latter alternative is chosen, the expenditure may well be of a capital nature.

 

27. Counsel further submitted that the revenue nature of Auckland Gas’s expenditure is shown or confirmed by the company's objective. The objective was not born out of a desire to improve the distribution system or to add new or improved features. The objective was to restore the system to its original functional and reliable state. The method adopted, of inserting polyethylene piping, happened to be the most effective and cheapest way to achieve that goal. The Court of Appeal, it was said, ignored the practical and business goal Auckland Gas set out to reach. Their Lordships cannot agree. As already noted, the desire to solve a maintenance problem is not inconsistent with carrying out work of a capital nature. The nature and extent of the work carried out to the physical asset are what is determinative of the character of the work. The fact that the method chosen is the cheapest and most effective is neutral. It does not deprive expenditure of its capital character. Replacing an object may be cheaper and better than patching and mending.

 

28. Reliance was placed on the decision of the (English) Court of Appeal in Morcom v. Campbell-Johnson [1956] 1 Q.B. 106. This was a landlord and tenant case. The landlords of a block of flats spent money on replacing an old two-pipes drainage system with a modern one-pipe system. Denning L.J. said (at page 115):-

"... the drainage system in these flats is the same now as the system which existed before. All that has happened is that, instead of there being two pipes to carry the water and refuse away, there is one pipe. That is simply the replacement of the older two-pipes system by its modern equivalent of one pipe. It comes, I think, properly within the category of repairs and not that of improvement. So, also, with the cold-water system .. In both cases, as the surveyor said, it was a cheaper way of doing it than it would have been simply to restore the old system as it was."

 

29. On its facts this decision may seem to afford support for the taxpayer. The appearance is deceptive. In the Morcom case the issue was whether the landlord was entitled to increase the rent payable by the tenant under the Increase of Rent & Mortgage Interest (Restrictions) Act 1920. That turned on whether the landlords had incurred "expenditure on the improvement ... of the dwelling-house (not including expenditure on decoration or repairs)". Whether work constituted an improvement was to be looked at from the point of view of the tenant: see Hodson L.J., at page 120. That is not the appropriate test in the present case.

 

30. Counsel submitted that the Court of Appeal's analysis emphasised physical form rather than physical function. He relied on a statement of Windeyer J. in W. Thomas & Co. Pty. Ltd. v. Commissioner of Taxation of the Commonwealth of Australia (1965) 115 C.L.R. 58, 72, to the effect that when considered from the point of view of use as distinct from appearance, it is restoration of efficiency in function of an object rather than exact repetition of form or material that is significant. Here again, their Lordships have to part company with the appellant's argument. As a general proposition Windeyer J.'s dictum is, no doubt, sound and helpful. It would be in point in this case if defective pipes or joints had been replaced or made good, even if in not exactly the same form or with exactly the same materials as before. But that is not what occurred. The approach of the Court of Appeal was correct.

 

31. For these reasons, which are substantially the same as those of the Court of Appeal, their Lordships will humbly advise Her Majesty that this appeal ought to be dismissed. The appellant must pay the respondent's costs before their Lordships' Board.