Deductions
1783. What is a `repair'?
November 2009 - Issue 123
Section 11(d) of the Income Tax Act, No. 58 of 1962 (the Act) provides that a taxpayer is entitled to a deduction in respect of expenditure incurred on "repairs" of property occupied for purposes of trade or which produces income.
But what exactly is a "repair"? The Act contains no definition, and it has fallen to the courts to interpret this elastic word.
The crucial distinction that has been drawn by the courts in this regard is between, on the one hand, "repairs" (expenditure which is deductible in terms of section 11(d) of the Act) and an "improvement, reconstruction or renewal" (expenditure which is regarded as being of a capital nature, and therefore not deductible). This capital expenditure may however qualify for a wear and tear or depreciation allowance in terms of section 11(e) of the Act, which allows qualifying capital expenditure to be deducted in installments over the estimated productive life of the item in question. This, of course, is second prize — it is far more valuable for a taxpayer to be entitled to deduct the whole of the expenditure, as a "repair", in the year in which it was incurred.
Repair presupposes a defect or malfunction
In Flemming v KBI [1995] (57 SATC 73), the Appellate Division said that our courts, as with the English courts, have not succeeded in laying down precise guidelines to distinguish between a ‘repair’ and an ‘improvement’.
The court went on to say that the common element of the various dictionary definitions was that they "refer to the rectification, refurbishing or repair of an object which, in comparison with its previous condition, had developed a defect or shortcoming".
In Flemming’s case, the taxpayer had a usufruct over a farm. The supply of water to an existing borehole on the farm, to which a windmill was attached, reduced substantially, apparently because of dwindling underground water and not because of any defect in the borehole.
At the insistence of the lessee of the farm, the taxpayer arranged to have a new borehole drilled some hundred metres from the first one, and the new borehole yielded adequate water. The taxpayer claimed the cost of drilling the new borehole and the cost of acquiring and erecting a new windmill over it as a "repair" in terms of section 11(d).
The court held that there was no evidence that the first borehole was faulty, thereby necessitating its replacement. It was held that the taxpayer had incurred the expenditure in question, not on "repairs of property", as contemplated in section 11(d), but on the improvement of the water supply. The expenditure was held not to be deductible.
A "repair" can involve replacement
The difficulty with the proposition, accepted by our courts, that a "repair" must be distinguished from an "improvement", is that "repair" inevitably results in an "improvement" to the property in question, in that what was defective has now been put right.
Nor is there a clear dividing line between a "repair" and a "renewal", as many repairs involve renewal; for example, a broken window pane is replaced with a new pane, and a malfunctioning machine component is replaced with a new component.
The unsatisfactory attempt to identify a (deductible) "repair" by contrasting it with a (non-deductible) "improvement" has driven the courts to try to differentiate between the two concepts on the basis of the proposition articulated in the seminal English case of Lurcott v Wakely and Wheeler in 1911, in which it was held that —
"Repair" and "renew" are not words expressive of a clear contrast. Repair always involves renewal; renewal of a part; of a subordinate part. . . 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."
This approach has been endorsed by the courts in South Africa. (See for example CIR v African Products Manufacturing Co Ltd [1944] (13 SATC 164).
The problem is that this approach merely exchanges one difficult question ("what is a repair?") for another, no less difficult question (namely, how to determine whether the property that has been put right is a "subordinate part" of a larger "entirety").
Thus, for example, it may be self-evident that, if a taxi-owner removes a malfunctioning spark plug from his taxi and installs a new one, the expenditure incurred is in respect of a "repair" (and qualifies for an outright deduction) because a spark plug is a subordinate part of a larger entirety, namely the vehicle as a whole, or at least the engine.
But the matter may become more difficult when one considers larger installations, such as a production line, comprised of a number of specialized machines, each with a separate function, that are configured sequentially to produce a particular product in an entirely automated process. For example the first machine takes in a sheet of metal and cuts out shapes, the second presses the shapes to the desired form, the third coats the products with a protective coating, the fourth dries the coating, and the fifth wraps and packs the products. If one of the machines is replaced, is this a repair, or is each machine an "entirety"?
The difficulty of distinguishing between a "subordinate part" and the "entirety" came to the fore in Rhodesia Railways Ltd v Income Tax Collector, Bechuanaland [1933] (6 SATC 225). In this case, Rhodesia Railways Ltd, which owned and operated a railway, including the railway line, had replaced rails and sleepers on 33 miles of a railway line whose total length was 394 miles.
Was this a "repair" (and thus deductible expenditure) or was it a reconstruction or improvement (which did not qualify for deduction as a repair)?
It was clear that if Rhodesia Railways had replaced one sleeper, this would have been a "repair". It was equally clear that if it had replaced the rails and sleepers over the entire 394 miles, this would not have constituted a "repair", but a "reconstruction" of the line.
So, at what point on the 394 miles of the track would the replacement of one more sleeper have caused the taxpayer to cross the threshold between a "repair" and a "reconstruction"? Fortunately, the court did not have to resolve this metaphysical question (Courts only have to deal with the facts of the matter before them), and the court held that where, in the circumstances of this particular case, the taxpayer had replaced 33 miles of sleepers and rails, this constituted a "repair", and the expenditure thus qualified for an outright deduction.
A subsidiary part is determined by its economic function
In Australia, the practice of the revenue authorities, in determining whether property is an entirety or a subordinate part, is to ask whether the property in question provides a useful function without regard to any other components; whether it is separately identifiable as a principal item of capital equipment and whether it is physically or functionally an inseparable part of a larger entirety.
It has been suggested that the question as to what is a "subordinate" part and what is an "entirety" must be answered by looking to the economic function of the relevant part within the structure of the particular taxpayer’s business.
Thus, in Samuel Jones & Co (Devonvale) Ltd v IRC [1951] 32 TC 513 the taxpayer, who carried on a business of processing paper, claimed as a deduction the cost of replacing an old chimney in its factory with a new one, and contended that the expenditure qualified for an outright deduction as a "repair", on the grounds that the chimney was a subsidiary part of the factory.
The judge agreed, holding that —
"the facts seem to me to demonstrate without a doubt that the chimney is physically, commercially and functionally an inseparable part of an entirety which is the factory. It is doubtless an indispensable part of the factory, doubtless an integral part, but none the less a subsidiary part, and one of many subsidiary parts, of a single industrial profit-making undertaking."
PricewaterhouseCoopers
