McGee, Andrew --- "Inherent defects and the repair covenant in commercial leases" [2007] QUTLawJJl 20; (2007) 7(2) Queensland University of Technology Law and Justice Journal 325

INHERENT DEFECTS AND THE REPAIR COVENANT IN COMMERCIAL LEASES


DR ANDREW MCGEE [*]



I INTRODUCTION

Does the covenant to repair in a commercial lease include an obligation to put right an inherent defect in the building, if that inherent defect causes the building concerned to fall into a state of disrepair? Or is it the case that, whoever bears the obligation of repair under the relevant covenant, that person need only remove the damage caused by the inherent defect, without removing that inherent defect itself? What happens if the only way the damage can be remedied is by removing the inherent defect? Does such work still constitute ‘repair’, or does it amount to ‘renewal’ and so fall outside the repair covenant?

There has been an abundance of case law on this topic,[1] yet the position, at least in Australia, remains unclear. Indeed, although the current position in the UK is clear, it will be argued here that much of subsequent judicial construction of the early cases in the UK is vulnerable to criticism, and that this might constitute a ground for taking an alternative approach in Australia, particularly in light of the fact that the UK position arguably leads to unjust results where the bearer of the repair obligation is the tenant.[2]

In the following section, the respective UK and Australian positions are briefly outlined.[3] I then review the UK position in more detail, focusing, in particular, on judicial construction of the early cases, to see if there has been an unnoticed change in the UK position, a position which takes itself to be founded on the early cases but one which, in fact, arguably departs from them. In the conclusion, I suggest that it might be better for Australia to take a different track.

II THE COVENANT TO REPAIR IN THE UK AND AUSTRALIA

A Repair and Renewal Distinguished

Before reviewing the authorities in detail (a task reserved for section 4), it will be convenient to state the principle adopted with respect to repair in the UK. The bearer of the obligation to maintain a building ‘in good condition’ or ‘tenantable repair’, or simply ‘repair’,[4] is not under an obligation to renew the whole or substantially the whole of the building.[5] If it were otherwise, he or she would effectively be giving the landlord back a different thing from that which was taken under the lease.[6] It is obvious enough, of course, that ‘repair’ inevitably involves some renewal, and that the distinction between repair and renewal must be drawn, not by reference to parts of the building, but by reference to the whole or substantially the whole of it, in order for the distinction between repair and renewal to acquire a grip and so be given a sense. If the distinction were to be applied to parts, rather than the whole or substantially the whole, the distinction between repair and renewal would be untenable, because repair inescapably requires some renewal. The application of the distinction is well illustrated by the comments of Buckley LJ in Lurcott v Wakeley and Wheeler[7] in the following passage:

A skylight leaks; repair is effected by hacking out the putties, putting in new ones, and renewing the paint. A roof falls out of repair; the necessary work is to replace the decayed timbers by sound wood; to substitute sound tiles or slates for those which are cracked, broken, or missing; to make good the flashings, and the like. Part of a garden wall tumbles down; repair is effected by building it up again with new mortar, and, so far as necessary, new bricks or stone. 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 entirety not necessarily the whole but substantially the whole subject matter under discussion.[8]

Whether the work required in a particular case constitutes renewal or replacement of the whole/substantially the whole (and so falls outside the repair covenant), or renewal or replacement merely of defective parts, and so repair, is a question of degree.[9] For this reason, the distinction between repair and renewal will in some cases be blurry – some cases will fall at the borderline. In such cases, it will be a matter of judgement. Perhaps the addition of the words ‘or substantially’ to ‘the whole’ make this even more so, introducing a broader range of different fact scenarios which might be decided differently, depending on the intuitions of the particular judge whose responsibility it is to decide the issue in a particular case. This has arguably led to some inconsistencies in application of the principle in the UK.[10] As noted by Antra Hood, the test is ‘deceptively simple’.[11] Perhaps an indication of the deceptive nature of this simplicity is the fact that no less than three different tests (all arguably differing only in verbal formulae) are to be found in the authorities. Mustill LJ expounds the tests in McDougall v Easington District Council[12] as follows:

(i) whether the alterations went to the whole or substantially the whole of the structure or only to a subsidiary part;
(ii) whether the effect of the alteration was to produce a building of a wholly different character from that which has been let;
(iii) what was the cost of the works in relation to the previous value of the building, and what was their effect on the value and lifespan of the building?

It is not my intention to analyse the ‘surprising results’ of the application of these tests referred to by Antra Hood.[13] Instead I shall focus merely on the implications of the tests for cases which have concerned what might be called inherent defects. It is here that there might arguably be a divergence of views between the UK and the Australian authorities.

B Repair and Inherent Defects – the UK Position

An inherent defect refers essentially to a defect in the building which is a product of improper design.[14] In one sense, of course, putting right an inherent defect might, at least in ordinary language, accurately be described as repairing the building. Add to this the fact that, in the UK, it has been said on more than one occasion that ‘repair is an ordinary English word’,[15] and one would be forgiven for thinking that inherent defects would naturally fall within the covenant to repair even though, strictly speaking, no damage has been caused to the building. This is not, however, the case. Precisely because, if there is no damage resulting from the inherent defect, the building cannot be said, in any meaningful sense, to have fallen into disrepair, the building cannot be said to be in need of repair.[16] In short, defective design in the building is not the same thing as damage to that building.[17] Rather, the building has, from its inception, existed in a defective state, and so has not subsequently fallen, from a perfect state, into one of disrepair. Consequently, removal of an inherent defect does not require ‘repair’ of the building. This proposition is settled law.[18] To that extent, it is perhaps misleading to claim that ‘repair’ is an ordinary English word, given that any lay person passing by a house where workmen might be putting right the inherent defect would undoubtedly, if asked what the workmen were doing, say that the workmen were ‘repairing’ the house. Far from ‘repair’ being an ordinary English word, it has a technical legal sense: it applies to a building which has fallen, from a state in which there were was no disrepair, into one in which there is disrepair, and which is therefore in need of repair. It does not apply to a building which, although of defective design which requires attention, has not fallen into a state of disrepair.

The point can be illustrated by the case of Quick v Taff-Ely Borough Council.[19] In that case, the plaintiff tenant leased a house which, over a number of years, suffered from severe condensation. The condensation caused bedding, furnishings, clothes and woodwork to rot, making living conditions in the house almost unbearable. The condensation was caused by an absence of insulation around the concrete window lintels and single-glazed metal-frame windows. Importantly, the liability under the repair covenant of the landlord local authority was to keep the structure and exterior of the house in repair, not the decorations. Consequently, although damage was caused to the decorations, in the relevant sense, the inherent defects in the house – the absence of insulation around the concrete window lintels and the single-glazed metal-frame windows – did not cause ‘damage’ to the building, but only affected the amenity of the premises. Consequently, the building was not in a state of disrepair, and the landlord local authority was not liable under the repair covenant. As Dillon LJ expressed the point:

the key factor in the present case is that disrepair is related to the physical condition of whatever has to be repaired, and not to questions of lack of amenity or inefficiency...Where decorative repair is in question one must look for damage to the decorations but where, as here, the obligation is merely to keep the structure and exterior of the house in repair, the covenant will only come into operation where there has been damage to the structure and exterior which requires to be made good.[20]

The case of Quick v Taff-Ely Borough Council[21] was applied in Post Office v Aquarius Properties Ltd.[22] In this case, the lease contained an obligation on the tenants ‘to keep in good and substantial repair...the demised premises and every part thereof’. There was a defect in the structure of the basement of an office building which was present from the time the building was constructed. In periods when the water table rose, the defect allowed ground water to enter the basement so that the water stood ankle deep on the floor for some years. However, aside from permitting water to enter the basement, no damage was caused to any part of the building by the defect. Accordingly, both at first instance, and on appeal, the tenants were found not to be liable under the repair covenant: the building had not fallen from a state of repair into a state of disrepair and the element of damage necessary to bring the repair covenant into operation was therefore absent.[23]

Thus, where there is an inherent defect in the building – for example, a defect in the construction of the building – but no damage is caused by that defect, the tenant (or landlord, as the case may be) is not liable, under the repair covenant, to remove the defect. The law on this point is the same, both in the UK and in Australia.

On the other hand, where the building has fallen into disrepair as a result of the inherent defect, that is, where the inherent faultiness in the design and/or construction[24] of the building has caused damage to the building – for example, defective foundations have caused the walls of the building to bulge and/or crack – the UK and Australian authorities diverge. In the UK, the inherent defect itself, and not merely the consequent damage thereon, may fall within the repair covenant.

The law in the UK is that, regardless of how the damage is caused, as long as there is damage, then the work required to put right that damage – even if putting right the damage includes removing the inherent defect that caused it – will fall within the covenant to repair, provided that the work required involves no more than renewal of defective parts of the building, rather than renewal of the whole or substantially the whole of the building. As we shall see in more detail below, in the case of Ravenseft Properties Ltd v Davstone (Holdings) Ltd,[25] Forbes J, reviewing the UK authorities in response to the tenants' submission that inherent defects do not fall within the repair covenant, stated:

[T]he court, when dealing with wants of reparation caused by an inherent defect, chose to treat the matter as one of degree, and in Brew Brothers v Snax (Ross) Ltd [1970] 1 QB 612, the court effectively said that every case, whatever the causation, must be treated as one of degree....The true test is, as the cases show, that it is always a question of degree whether that which the tenant is being asked to do can properly be described as repair, or whether on the contrary it would involve giving back to the landlord a wholly different thing from that which was demised (emphasis added).[26]

The argument that the damage was caused by an inherent defect in the building and so should not be the tenant's responsibility under the repair covenant has been strongly and consistently rejected. One simply looks at the work required to put right the damage and remove the defect, applying one or more of the three tests referred to by Mustill LJ in McDougall v Easington District Council[27] and, depending on the outcome of the application of that test, it may be that the bearer of the repair obligation under the covenant must remove an inherent defect in addition to the damage caused thereby.[28]

Thus, on the UK position, it is not possible to state unqualifiedly that inherent defects do not come within the repair covenant at all. Whether or not an inherent defect comes within the repair covenant will depend on two factors: whether damage is caused and, if so, the extent of the damage, and therewith the extent of the work required to put right that damage. It is on this point – where the inherent defect causes damage – that the UK and Australian positions appear to diverge. I will examine these UK authorities in more detail in section 4 of this paper.

C Repair and Inherent Defects – the Australian Position

The Australian position seems to differ from that taken in the UK where damage occurs as a result of the inherent defect. In Lazar v Williamson and Others,[29] during the tenant’s lease of a theatre, it became necessary to replace the defectively constructed beams in the roof of the theatre in order to make the roof secure so that the premises could continue to be operated as a theatre. The beams were so constructed that, ‘when machinery used for the purpose of the theatre was set up, the beams sagged, bent downwards, and lost the rigid condition necessary for the security of the roof.’[30] The defective construction had caused the theatre to fall ‘into a condition of dilapidation or disrepair’.[31] In other words, the inherent defect had caused damage to occur. Sir J Martin CJ and Faucett J, with Sir G Innes J dissenting, held that the work required was not within the repair covenant because it was a structural defect. Sir J Martin CJ said:

It appears to me that the parties to this lease, that is to say, Samuel Lazar and the defendants, when they entered into this sub-lease, did not in any way have in contemplation the expending of any money for making good any structural defects in the theatre. The building was not constructed in the substantial way in which it ought to have been.... We cannot gather that these parties in any way intended that, under this covenant to repair, any alterations in the theatre were contemplated...I am of the opinion that the covenant to repair does not apply to any alteration or reconstruction of the building, either wholly or in part. The alterations made do not come within the meaning of the word ‘repairs’.[32]

According to Sir J Martin CJ, then, liability for structural defects which require addressing does not come within the contemplation of the parties on entry into the lease. Faucett J was of the same opinion:

As to liability of the defendants under th[e] covenant, the law is too clear to admit of question. No doubt, if at the time of the lease the house is out of repair, the lessee, under this covenant to repair, is bound to put it in repair...The question, then, is whether the repairs executed were such as were contemplated by the parties to the covenant sued upon. If such repairs were rendered necessary by ordinary wear and tear, the defendants would be liable; but they would not be liable if, by any structural defect in the building, the repairs needed are not such as were contemplated by the parties who entered the contract.[33]

Sir G Innes J dissented on the ground that once the building has fallen into a state of disrepair, as happened here, the work came within the repair covenant notwithstanding that the disrepair was caused by inherent defects: ‘I cannot see that dilapidations so caused do not come within the covenant to repair’.[34] His Honour therefore adopted the position that has subsequently become the orthodox position in the UK.

In Graham and Anor v The Markets Hotel Pty Ltd,[35] Jordan CJ held, for reasons that do not clearly emerge, that the removal of lavatories was not a breach of the obligation to yield up the premises in good repair because the original lavatories were defectively constructed, and there was no obligation on the tenant to make good inherent defects. In a significant passage of the judgement in the New South Wales Court of Appeal, Jordan CJ made the following remark:

if, when it was taken over, [the structure] was subject to some inherent defect of a substantial kind, an agreement merely to repair does not impose an obligation to remove the defect, but only to maintain the structure subject to the defect so far as this can be effected by repair. By inherent defect is meant some original or supervening defect of an abnormal kind, such as would not be found in a properly built structure, would not be produced in such a structure by the degenerative processes of user and decay, and cannot be remedied except by the replacement or remodelling of the structure or some substantial part of it.[36]

On appeal to the High Court, it was decided that, on the facts, the case did not concern an inherent defect. But it is noteworthy that, in obiter, Latham CJ endorsed Jordan CJ’s comments concerning the absence of liability under a repair covenant for inherent defects. Latham CJ cites Jordan CJ’s comments with approval.[37] Importantly, however, Latham CJ also says:

There is a difference between repairing a house and building a new house in place of an old house. It is a question of degree whether rebuilding part of a house does or does not fall within the category of repairing a house. A covenant to repair does not involve the covenantee in an obligation to make improvements, but if he cannot perform his covenant to repair without making improvements, then the expense of making the improvements falls upon him.[38]

These comments mirror similar comments made by Jordan CJ in the Court of Appeal, where his Honour, just before making the remarks concerning inherent defects, stated:

Under an ordinary agreement to repair, there is no obligation to replace the structure wholly or substantially if, without the fault of the party agreeing, replacement of this kind becomes necessary. But, as a general rule, the fact that the structure has fallen into such a state of disrepair that the necessary repairs can be effected only by the replacement of part – even a considerable part – of the structure, does not absolve a party from his liability to repair.[39]

These comments of Latham CJ and Jordan CJ, together with the claim that a tenant is not liable to put right an inherent defect but only to maintain the structure subject to the defect so far as this is possible by repair, show that their Honours effectively espouse two separate principles. They agree with the UK authorities that it is a question of degree whether the work required is repair or renewal. But they disagree that this is the only relevant consideration at issue in all cases. On the contrary, when there is an inherent defect causing damage, the tenant is never liable to put right the inherent defect but only to maintain the structure, so far as is possible, subject to the inherent defect. Inherent defects do not fall within the covenant to repair at all.

The difference in these two principles is expressed even more clearly by Nevile J in Clowes v Bentley Pty Ltd.[40] The case did not concern an inherent defect, but his Honour, having reviewed the authorities, stated the following qualification:

I should perhaps add the qualification that the plaintiffs were under no obligation to put right an inherent defect in the premises which would inevitably at some future time either within the term of the lease or after its expiration cause the collapse or destruction of the premises, nor would their obligation extend to renewing substantially the whole of the demised premises, which would in effect give to the lessor a different thing than that which the tenant took, as distinguished from the renewing of subsidiary parts which must be involved in the connotation of repair.[41]

As we saw in the case of Graham and Anor v The Markets Hotel Pty Ltd in both the Court of Appeal and the High Court,[42] Nevile J here clearly formulates two principles, only the second of which is identical to the position taken in England. The first principle is that the tenant is under no obligation to remove an inherent defect that would result in damage to the property – even if that damage is substantial enough to result in the collapse or destruction of the premises.[43] The second principle is that the obligation does not extend to renewing the whole of the demised premises, as distinguished from renewing subsidiary parts. This is the principle that coincides with the UK position. That his Honour nevertheless treats these as two separate principles is clear enough from his use of the word ‘nor’: ‘nor would their obligation extend to renewing substantially the whole of the premises.’ Consequently, in these cases, the presence of an inherent defect of a substantial kind cannot be seen to be catered for by the principle that the obligation to repair does not extend to renewing the whole of the demised premises.

These cases notwithstanding, the Australian position is not, however, completely clear-cut. The Tasmanian case of Martin v Croft,[44] seems to accord with the English authorities. The case concerned damage caused by drains that were defective prior to the entry into the lease. Interestingly, none of the Australian authorities are discussed by Wright J in reaching his decision. Instead, he relies exclusively on the distinction between repair and renewal, and the English authorities of Lurcott v Wakeley and Wheeler[45] and Brew Bros Ltd v Snax (Ross) Ltd,[46] together with the orthodox interpretation given in those cases of Lister v Lane and Nesham,[47] about which we shall have more to say below. Both Lurcott and Brew did not concern inherent defects, so it is unsurprising that his Honour would find in those cases only the principle that a tenant is not obliged to renew the whole of the premises, and that anything falling short of such renewal will be repair, and therefore within the repair covenant. For example, as we see below, Lurcott concerned only the operation of time on a wall; it did not concern the operation of time on a defectively constructed wall. The notion of an inherent defect did not arise on the facts and consequently was not discussed. His Honour was not, of course, bound to follow Graham or Clowes, these being cases from other Australian jurisdictions. But these cases are nevertheless of persuasive force and should arguably have been addressed as being more on point.[48]
Before seeing whether there originally were two distinct propositions in the UK case law by carrying out a more thorough review, we shall next look at whether the UK position as it currently stands is capable of yielding unjust outcomes.

III IS IT POSSIBLE THAT THE UK POSITION MIGHT LEAD TO INJUSTICE?

There can be no doubt that the position as articulated in the UK has not been considered by any of the judges who have turned their mind to the issue to give rise to any potential injustice. Nevertheless, the difficulty with the position can be shortly stated. Effectively, where damage is caused by a design fault in the building then whether or not the repair covenant has been broken has been made to depend simply on the extent of the damage, or the amount of work required. If the damage is minor, and the extent of the work required does not amount to renewal of the whole or substantially the whole of the building, the tenant in failing to remove the damage and the defect which caused it will have breached the covenant to repair. If, on the other hand, the damage is major and cannot be put right without renewal of the whole, then the tenant will not be liable.

But why should a breach, and therefore a responsibility for damage, be made to depend entirely on the extent, rather than the cause, of the damage? It seems very strange indeed that legal responsibility should vary simply with the extent of work required, without any regard whatsoever to the cause of the damage which gave rise to the need for the work in the first place. The argument that the overall ratio of cost – the third of the tests outlined by Mustill LJ in McDougall v Easington District Council[49]– might alleviate the injustice is untenable. If the breach is caused by something beyond the contemplation of the parties when entering the lease,[50] it is no answer to the tenant to say to him that, be that as it may, the costs of removing the defect are minimal, so the tenant should be responsible.

The essential point here is that the risk of an inherent defect becoming manifest during the term of the lease – a defect that would not be discoverable at the time the lease was entered into – would not be something that a tenant would reasonably contemplate having responsibility for under the repair covenant, as it is something completely beyond their control. True it is that there are, of course, other risks that can equally be said to be beyond the control of a tenant, for instance, a passing car on a newly gravelled street might flick the gravel onto a window and cause a crack in the window, and the tenant would, of course, be liable under the repair covenant for the repair of the window. But these kinds of risks are ones that arise from occupying the premises during the term of the tenancy. The tenant does not, however, have any control over the construction of the premises they come to occupy, and should not therefore have to bear the risk of the building becoming defective if that defect was not discoverable at the time the lease was entered into. To hold otherwise is to make the tenant responsible for something that happened prior to the lease and over which he had no control. The landlord is normally in a superior bargaining position and can make it the tenant’s responsibility by providing a covenant to that effect in the lease. It seems wrong for the courts to go further than the lease itself goes by unnaturally construing the repair covenant to make the removal of an inherent defect causing damage ‘repair’ for the purposes of that covenant.

A second difficulty with the UK position is a logical one. As we have seen, the rule is that an inherent defect is not itself damage, and so is outside the repair covenant until damage results from it. If someone puts it right even though there has been no damage, then they cannot be said to be repairing the building. This holds even if the defect is minor, not only if it is major – the point is that the defect is not itself damage and so does not amount to disrepair. Yet when damage is caused by the inherent defect, if that damage is only minor, and the defect causing it is minor, the defect suddenly comes to fall within the repair covenant, as happened in Ravenseft. This is an illogical state of affairs. As a matter of logic, only the damage, and not the inherent defect, should fall within the repair covenant, because an inherent defect is not itself damage – and does not, therefore, amount to disrepair.

It might be responded that, be that as it may, if removal of the damage requires removal of the defect, and removal of the damage is within the repair covenant, then to comply with that covenant, the tenant must remove the inherent defect when removing the damage in order to prevent that damage being caused again, etc. But such a reply begs the question. The very point at issue is that, when the damage is itself the result of the inherent defect, and the inherent defect, not being damage, is outside the repair covenant, it is illogical to hold the tenant liable under the repair covenant for the inherent defect as well as for the damage caused by it, simply on the ground that the damage cannot be remedied without removal of the inherent defect. To hold otherwise is to blur the distinction between disrepair and inherent defects, the very distinction that the law has consistently upheld as a necessary and cogent one, and it is no answer to this logical point to say that, nevertheless, if the resultant damage cannot be removed without the removal of the inherent defect, that takes the inherent defect inside the repair covenant. On the contrary, it is more rational to conclude the very opposite: the fact that the damage is caused by an inherent defect, and an inherent defect is outside the repair covenant, means that the damage resultant thereon should likewise fall outside the repair covenant, if the only way to remedy the damage is to remove the inherent defect.

If the reason inherent defects are not disrepair is that the premises are in the same condition as they were in when they were taken (no damage) then there is no reason why, when those defects themselves start to cause actual damage, they should then be classified as disrepair (rather than the actual damage they cause being so classed).

An alternative analysis would be to say that the tenant is only liable for the damage resulting from the inherent defect, and not the inherent defect itself (rather than to say, as we just have, that the tenant should be liable neither for the damage resulting from the defect nor the defect causing the damage). This position is, of course, maintainable in a situation where the damage can adequately be remedied without removing the inherent defect. But what about situations where the damage can be remedied only for a short time, before it recurs again owing to the inherent defect? For the following reasons, this analysis ends up being identical to the one that I have just been defending: if the tenant were to be made liable only for the damage resulting from the defect, rather than the defect itself, this would lead to the absurdity of the tenant repairing the damage only to see it recur, repair it again, see it recur again, and so on and so forth. A court would not regard a patch up job of this sort as constituting adequate ‘repair’ for the purposes of the repair covenant. As a matter of common sense, if the damage caused by the defect is going to recur within a short space of time, the work undertaken to remedy the defect would not be properly be considered to have reached the standard of repair that would be required under the covenant. This would therefore be a situation in which the damage caused by the inherent defect cannot be removed without removing the inherent defect.

The above analysis of the difficulties besetting the UK position no doubt partly explains why a different position – perhaps with one or two exceptions[51] – is taken in Australia, where damage that is caused by an inherent defect does not fall within the repair covenant. It is submitted that this analysis is more persuasive than that espoused in the UK courts, and should remain the position in Australia should the issue be litigated and reach the highest level here.

In the following section, I attempt to strengthen this conclusion by reviewing the English authorities in more detail. We shall see that, initially at least, there was arguably separate recognition given to cases where inherent defects were concerned, but that this separate recognition was eventually eroded through judicial interpretation of those earlier cases. If some of this interpretation might respectfully be thought to be erroneous, then, to that extent, such a conclusion would provide additional grounds for adopting a different analysis in Australia should the issue be litigated again here, particularly if the litigation should reach the High Court.

IV REVIEW OF THE ENGLISH AUTHORITIES

In this section, a detailed review of the English authorities is undertaken. I will approach this task by taking as my primary focus the influential judgement of Forbes J in Ravenseft Properties Ltd v Davstone (Holdings) Ltd,[52] in which his Honour undertakes an examination of the authorities. I propose to review Forbes J’s own consideration of the authorities here to see if his analysis is justified. Although his Honour’s judgement has been approved on more than one occasion by the English Court of Appeal and has been described by that court as a ‘careful’ judgement,[53] it will be seen that some of his Honour’s conclusions from the authorities are vulnerable to criticism.

A Ravenseft: the Facts and Decision

In Ravenseft, a 16-storey block of maisonettes was constructed of a reinforced concrete frame with stone cladding. Expansion joints were omitted from the structure because at the date of construction it was not standard practice to include expansion joints – it was not known, at the time, that the expansion rate for the concrete frame differed from that for the stone cladding. About 13 years after construction, part of the stone cladding became loose and in danger of falling, owing to the bowing of stones caused primarily by defective design, though the danger was also partly caused by defective workmanship in failing to secure the stones. The landlord effectuated the necessary remedial work and claimed the cost of doing so from the then tenant under the repair covenant. The tenant denied liability under the covenant, on the ground that the damage was caused by an inherent defect and defective workmanship. The total cost of the work was £55,000 of which only £5,000 was attributable to the work of inserting the joints. The cost of erecting the building in 1973 would have exceeded £3 million.

Forbes J held that the tenant was liable under the repair covenant. Reviewing the authorities, he rejected counsel’s proposition that there was a doctrine that want of repair due to an inherent defect fell outside the repair covenant. Rather, it is a question of degree whether that which the tenant is being requested to do amounts to repair, or alternatively would result in the tenant giving back to the landlord a wholly different thing from that demised, and so amounts to renewal, thereby falling outside of the repair covenant.[54] Alternatively, applying the test of the costs of the works in relation to the value of the building, Forbes J stated that he could not ‘accept that the cost of inserting these joints could possibly be regarded as a substantial part of the cost of the repairs, much less a substantial part of the value or cost of the building.’[55]

B Ravenseft: the Law – Forbes J’s Review of the Authorities

In Ravenseft, counsel for the tenant had argued that there was a ‘doctrine that want of repair due to an inherent defect could not fall within the repair covenant’. The contention was rejected by Forbes J. Instead, his Honour said, the law has consistently maintained that the question to be posed is simply whether what the tenant is asked to do can be fairly represented by the word ‘repair’, a question that is to be judged as a matter of degree in each case. His Honour cites what is perhaps the most well-known and often-cited passage in support of this proposition, from Lord Esher MR in Lister v Lane and Nesham:[56]

if a tenant takes a house which is of such a kind that by its own inherent nature it will in course of time fall into a particular condition, the effects of that result are not within the tenant's covenant to repair. However large the words of the covenant may be, a covenant to repair a house is not a covenant to give a different thing from that which the tenant took when he entered into the covenant. He has to repair that thing which he took; he is not obliged to make a new and different thing, and, moreover, the result of the nature and condition of the house itself, the result of time upon that state of things, is not a breach of the covenant to repair.

This passage has consistently been cited as support for the proposition in other cases.[57] It is taken to confirm the point that whether or not the work required falls into the covenant to repair is to be determined by reference to whether what is required would involve the tenant giving back to the landlord a new and different thing from that which he took under the tenancy. If it does, then the work required falls outside the repair covenant, and the tenant is not liable to perform the work under that covenant. Otherwise the tenant is liable. In other words, the passage is authority for the proposition that renewal of the whole or substantially the whole of the building is not repair. No doctrine that want of repair due to an inherent defect cannot fall within the repair covenant appears in the passage.

The problem with this – admittedly authoritative (given the number of times it has been cited as authority for the proposition endorsed by Forbes J) – interpretation of the passage is simply that it is incapable of accounting for the first sentence that appears there. In that first sentence, Lord Esher MR states without qualification that, ‘if a tenant takes a house which is of such a kind that by its own inherent nature it will in the course of time fall into a particular condition, the effects of that result are not within the tenant’s covenant to repair’ (italics added). The subsequent judicial gloss on the passage is not capable of explaining the meaning of this sentence. On the contrary, according to that subsequent judicial interpretation, the fact that the damage might have been caused by an inherent defect is completely irrelevant. In applying the test of whether the work required is renewal or whether it can properly be termed ‘repair’, no reference is necessary to the notion of an inherent defect – it completely drops out of account as immaterial to the issue. The difficulty with this interpretation of the passage, then, is simply that it cannot explain why Lord Esher MR made reference to a condition caused by the building’s own inherent nature, nor the unqualified nature of Lord Esher MR’s proposition that the effects of that result are not within the tenant’s covenant to repair. Forbes J’s (and earlier and later judicial) failure to explain the first part of the passage is therefore a consideration that counts against this interpretation of Lister.

One response to this criticism might be that the second sentence explains what Lord Esher MR must have meant by the first one. The argument would be that Lord Esher MR explicitly states that the covenant to repair a house is not a covenant to give back a wholly different thing from that which the tenant took when he entered into the covenant. But the problem with this response is two-fold. First, if this argument were correct, it would follow that Lord Esher MR had himself fallen into error. For rather than explaining the first sentence, the second sentence would, on this interpretation, falsify the first sentence: if the issue is simply whether the work required is repair or renewal, then it would be false to say that, if a building is of such a kind that by its own inherent nature it falls into a particular condition, the effects of that result are not within the covenant to repair. On the contrary, whether or not the effects are within the covenant to repair would depend on whether or not their remedy would require renewal, or simply repair, regardless of the inherent nature of the building.

The second problem with the response is that an alternative explanation for the second sentence has not been ruled out – an explanation which is, moreover, capable of making the passage read consistently and which is, therefore, a superior interpretation from that adopted by Forbes J and other authorities. It is submitted that there are two reasons why one might be giving back to the landlord a different thing from that which the tenant took under the tenancy. The first is indeed because one might undertake a renewal of the whole or substantially the whole of the building. But a second, different, reason might be because one is removing an inherent defect from the building. When a building has been defectively constructed, then it is from its inception an imperfect structure, even though no damage may yet have manifested itself as a consequence of the defective design or construction.[58] A building without that defect – a building that has not been constructed defectively, or is not of a defective design – is necessarily a different thing from one which has been constructed defectively, or is of a defective design. This is so regardless of whether the work required to remove the defect requires renewal of the whole or not. This is because, if the building is defectively designed or constructed, it has never been anything other than defective in its nature. In particular, as we know from the discussion above, it has not fallen from a state in which the defect was absent to one in which the defect is present. This makes it completely unlike a case of disrepair. If disrepair, as opposed to an inherent defect, is removed, then the building is restored to what it was originally. In the case of an inherent defect, by contrast, the tenant takes what was from its origin and by its very nature a defective building. If the tenant removes the defect, the tenant does not restore it to what it was originally, but rather makes of the building a different thing from that which it was originally. Therefore, if the tenant were to remove the defect, then even if the removal of the defect would not require the renewal of the whole or substantially the whole of the building, the tenant would nevertheless be returning to the landlord a different thing from that which he took under the tenancy.

In short, the reference to ‘giving back a different thing’ can refer to two different scenarios:

1. it can refer to the more usual scenario where the extent of work required on a building might amount to renewal rather than repair. In such a case, if the tenant were liable for the work, he would be giving the landlord a different thing from that which he took under the tenancy; but it can also refer to:
2. the difference between a defective building and a building that is not defective. These are two different things. If the tenant takes a defective building and returns a building that is not defective, he or she has given back to the landlord a different thing from that which was taken under the tenancy – he or she has, in essence, given back a building that is different in kind from that which was received under the tenancy.[59]

This interpretation of Lord Esher MR’s passage is capable of rendering the first two sentences of the passage consistent with one another. Recall that, in that famous passage, Lord Esher MR expresses the proposition in the first sentence in an unqualified way; he says simply that, if the tenant takes a house which is of such a kind that by its own inherent nature it will in the course of time fall into a particular condition, the effects of that result are not within the tenant’s covenant to repair’.[60] He does not say: it may not be within the covenant to repair, depending on whether what is required to be done amounts to renewal or repair of the whole. If the tenant removes an inherent defect, then necessarily something other than what was received is given back, regardless of the extent of the work required.

Further support for this interpretation of the passage can be found if the passage is read in its context. Lord Esher MR explicitly states: ‘You have to consider not only what the damage is – what is the amount of repair required – but also whether the covenant has been broken [italics added].’[61]

On Forbes J’s test, by contrast, one has to see what amount of ‘repair’ is required in order to see whether the covenant is broken: if the amount of work does not require renewal of the whole/substantially the whole, then the covenant is broken; otherwise it is the landlord’s responsibility. The criterion for whether the covenant is broken is therefore the extent of the work required. But this is not what Lord Esher MR says. On the contrary, although one considers the amount of ‘repair’ that is required, one also has to consider whether the covenant has been broken.[62] So for Lord Esher MR, there are two separate issues here. If, for example, the damage is caused by an inherent defect, then regardless of the amount of repair required, the tenant will not be liable for it under the covenant because the covenant will not have been broken. As such, the test Forbes J purports to find in the authorities, if that is indeed the test, departs from the authority of Lister.

Finally, support for this interpretation of the passage can be found if more attention is given to the third sentence cited in the famous passage. In that sentence, Lord Esher MR states: ‘He has to repair that thing which he took; he is not obliged to make a new and different thing, and, moreover, the result of the nature and condition of the house itself, the result of time upon that state of things, is not a breach of the covenant to repair (italics added)’.

The italicised portion of the quotation has received insufficient attention in judicial construction of the passage from Lister. It may be that, in this portion of the passage, Lord Esher MR is referring exclusively to the result of time on the building. But alternatively, the passage can be read as referring to both the result of the nature and condition of the house itself and the result of time on the building. If that is the case, then again he is explicitly stating, without qualification, that the result (ie damage caused by) the inherent nature and condition of the house itself is not a breach of the covenant to repair. Once again, there is no caveat that the result may not be a breach, depending on whether its removal requires repair or amounts to renewal.

If these arguments concerning the meaning of Lord Esher MR’s passage in Lister are correct, why is it that their meaning has been lost in subsequent judicial interpretation of them? I would suggest that their meaning has been lost because the same verbal formulation – ‘whether what is being given back is a different thing from what was given under the tenancy’ – is used to express what are essentially two different tests, one of which is used to determine whether the work requires putting right an inherent defect, and the other of which is used to determine whether the work required involves not merely repair, but also renewal, of the whole or substantially the whole of the building. For this reason, the test for an inherent defect became buried, so to speak, under the test for renewal of the whole. Another reason why the two different issues became merged is no doubt due to the fact that, on the facts in Lister, under the renewal test, the tenant would also be giving back to the landlord a different thing from that which he took under the tenancy, because removal of the defect in that case did, as a matter of fact, require a renewal of the whole building. This has, it is submitted, consequently misled subsequent judges in their consideration of the case into thinking that the only issue – regardless of whether there is an inherent defect or not – is whether the work required involves renewal of the whole or substantially the whole.

The next case considered by Forbes J is Southeby v Grundy.[63] In that case, the main walls of the house had been built entirely without footings, or at least on footings which were defective. The house was pulled down after being condemned by the London County Council as a dangerous structure under the London Building Acts, after the walls had started to bulge and had become fractured. The landlord sued the tenant for breach of the repair covenant. It was found, on the evidence, that the only way in which the order could have been avoided would have been by underpinning, which would have meant shoring up the premises, the removal of existing foundations, stage by stage, and the substitution of a new foundation in the way of footings and concrete. Lynskey J found that this work would have amounted to a renewal of the whole building, and so could not fall within the repair covenant.

Forbes J’s reliance on that case may, however, be questionable, because Lynskey J’s judgement is itself arguably open to objection. In Southeby Lynskey J, noting that he is concerned with the case of an inherent defect, reasons as follows:

That raises the question whether it can be said that this was a building which, by its own inherent nature would, in course of time, fall into a particular condition. I am sure that that is so, but the matter does not end there, because, in view of some of the other authorities to which I have been referred, particularly Anstruther-Gough-Calthorpe v McOscar, it seems to me that it must be a question of degree in each case.[64]

Forbes J cites this statement as authority for the proposition that there is no doctrine that inherent defects do not fall within the covenant to repair. Even if the case concerns an inherent defect, the question remains whether the work required involves giving the landlord something different in kind from what was demised (that is, whether the work is renewal or repair), and that question is a matter of degree in each case. But Lysnkey J may have erred by taking this view on the authority of Anstruther-Gough-Calthorpe v McOscar,[65] because that case did not concern an inherent defect at all; it concerned the standard of repair required under a repair covenant. The issue in that case was whether the tenants were liable to execute such repairs (taking account of the age, character, and locality of the premises) as would make the premises reasonably fit to satisfy the requirements of reasonably minded tenants of the class that would then be likely to occupy them, or whether, on the contrary, the tenants were liable for the costs of all acts necessary well and sufficiently to repair the premises so as to put them in that state of repair which they would be in were they managed by a reasonably minded owner. It was found that the latter was the requisite standard.[66] No issue concerning inherent defects was raised, or discussed, in the case. It is instructive to note that Anstruther’s case is the only case cited by Lynskey J to support his judgement. Given that he relies exclusively on the one case, and that the case relied on is off point, Southeby’s authority for the issues determined in Ravenseft must be considered questionable. It is noteworthy that Forbes J himself might be seen to acknowledge a difficulty with the judgement, when he comments that Lynskey J’s judgement: ‘is clearly to reject, or overlook, any argument that a want of reparation caused by inherent defect could not in any circumstances be within the ambit of the repairing covenant (italics added)’.[67]

It should be noted that to reject an argument, and to overlook an argument, are two completely different things, and Forbes J has not ruled out the possibility that Lynskey J has merely overlooked, rather than rejected, the argument. His reliance on Lynskey J’s judgement is for this reason unsound. To that extent, it is submitted the authority of Lynskey J’s judgement in Southeby must be considered questionable on this issue.

Forbes J then considers the case of Collins v Flynn.[68] In that case, inadequate foundations caused a pier carrying one end of a girder supporting a large portion of the back wall and, indirectly, part of the side wall of a house to subside, in turn causing fractures in the walls. The pier and the walls had to be rebuilt with newly designed foundations. Sir Brett Cloutman VC was concerned, in particular, with whether the tenant would be liable under a repair covenant for an inherent defect that affected, not the whole of the building, but only a subsidiary part of it. For this reason, the case is most interesting for our purposes. In that case, Sir Brett Cloutman notes that many of the cases in which the principle that a tenant will be liable for improvements necessary to subsidiary parts did not concern inherent defects. For example, Lurcott v Wakeley and Wheeler,[69] a leading case on the distinction between repair and renewal, did not concern an inherent defect,[70] thus the principle did not necessarily have any implications for the facts before him. To this extent, his Honour treated this case as less relevant for the determination of the issue in that case. On the other hand, the cases of Pembery v Lamdin,[71]and Wright v Lawson,[72] were significant because those cases did concern inherent defects and, in particular, defects affecting only subsidiary parts of the buildings in question, yet the tenant was not held in those cases to be liable under the repair covenant. Wright v Lawson,[73] for instance, concerned a bay window which had been erected and cantilever beams which were wholly inadequate and which caused the window to become dangerous. His Honour stated: ‘I observe that that this is no doubt a case of inherent defect in construction and relates only to a subordinate part of the structure’.[74]

Sir Brett Cloutman VC’s point here is that the case is treated differently from other cases – such as Lurcott v Wakeley and Wheeler,[75] – because, unlike in that case, Wright v Lawson,[76] concerned an inherent defect. To that extent, although Wright v Lawson was a case concerning only a subsidiary part of the structure, the decision there (that the matter did not fall within the repair covenant) was reconcilable with the view taken in Lurcott v Wakeley and Wheeler,[77]which held that renewal of subsidiary parts nevertheless falls within the covenant to repair.
Pembery v Lamdin,[78] the other case mentioned by Sir Brett Cloutman VC, concerned, more unusually, the question of whether the landlord (not the tenant) would be responsible under the repair covenant. In that case, the tenant took a lease of a ground floor shop and basement. She intended to use the basement as a cocktail bar. Owing to the age of the brickwork, but also because no steps were taken to waterproof the building when it was built, the building fell into a state of disrepair. Waterproofing of the external walls was required to remedy the defect causing the damage, and to render the basement usable as a cocktail bar. Although a substantial amount of work was required in order to effectuate the waterproofing, only a subsidiary part, and not the whole, of the building required work.[79] Despite this, the repair covenant was held not to apply, and the landlord was not liable to remedy the defect.

It is noteworthy, however, that the leading judgement in Pembery made extensive use of Lurcott v Wakelely and Wheeler,[80] which undoubtedly causes confusion for subsequent judges in so far as, strictly speaking, that case is not relevant to the issue of inherent defects, as Sir Brett Cloutman VC himself rightly notes.[81] This is undoubtedly the reason why Forbes J, when he himself considers Pembery v Lamdin in his judgement in Ravenseft, did not consider that the case posed any problems for his own decision.

Forbes J’s comments on Sir Brett Cloutman VC’s judgement in Collins v Flynn,[82] are instructive. This case represented a difficult authority for Forbes J, in so far as Sir Brett Cloutman VC held, as we have seen, that there could be no liability for an inherent defect, even where that defect concerned a subsidiary part of the building. This seems to confirm the interpretation of Lister’s case for which I argued above, where the unqualified nature of Lord Esher MR’s comments in the first sentence of the famous passage, discussed above, was taken to point towards the possibility, as we have seen, that ‘giving back a different thing’ might have two different interpretations – the important, and perhaps overlooked one being, for our purposes, that giving something back that is no longer inherently defective is ipso facto to give back something different in kind from that which was received under the tenancy, notwithstanding that the extent of the work required may not constitute renewal of the whole building.

How could Forbes J reconcile Collins v Flynn with Southeby v Grundy? He could not. In order to justify his decision in Ravenseft, Forbes J had to reject the analysis of Collins, and decline to follow the case. He does so by rejecting Sir Brett Cloutman VC’s interpretation of Southeby v Grundy in Collins. In Southeby, Lynsky J made the following remark:
It may be that the inherent nature of a building may result in its partial collapse. One can visualise the floor of a building collapsing, owing to defective joists having been put in. I do not think Lister v Lane would be applicable to such a case. In those circumstances, in my opinion, the damage would fall within the ambit of the covenant to repair, but, as I say, it must be a question of degree in each particular case.[83]

This comment in Southeby might be seen to represent a problem for Sir Brett Cloutman VC’s analysis, in so far as it states that if the inherent nature of a building resulted only in its partial collapse, Lister would not be applicable in that event. However, Sir Brett Cloutman VC interprets the remark as an obiter comment, and therefore did not consider himself to be bound by Lynsky J’s remarks in this passage. This is because Southeby was not concerned with what might be termed subsidiary parts of a building:

Plainly the doctrine of liability for the defects in a subsidiary part could have nothing to do with [Southeby’s] case. The case, as it seems to me, was on all fours with Lister v Lane and Nesham. Oddly enough, Lynskey J does introduce it, in what I think is an obiter passage. He said ([1947] 2 All ER at pp 761, 762): "It may be that the inherent nature of a building may result in its partial collapse. One can visualise the floor of a building collapsing, owing to defective joists having been put in. I do not think Lister v Lane would be applicable to such a case. In those circumstances, in my opinion, the damage would fall within the ambit of the covenant to repair, but, as I say, it must be a question of degree in each particular case." These "obiter joists", if I may so describe them, were mentioned both in Lister v Lane ([1893] 2 QB at p 216) and in Lurcott v Wakely ([1911-13] All ER Rep at p 44; [1911] 1 KB at p 914).[84]

Having treated the remarks of Lynskey J in Southeby as obiter comments, Sir Brett Cloutman VC concludes that the tenant was not obliged to put right the inherent defect in Collins:

I now come to the crucial point. Do the words "repair" and "renew" import a liability to rebuild with newly designed foundations and footings the pier supporting the girder, which in turn carries a great part of the rear wall and a part of the side wall in addition? This is manifestly a most important improvement, which, if executed by the tenant, would involve him in rendering up the premises in different condition from that in which they were demised, and on the authority of Lord Esher MR in Lister v Lane and Nesham, I do not think that the tenant is under any such obligation. Furthermore, although a suggestion of liability for removal of an inherent defect in a subsidiary part seems to have been touched on in Southeby v Grundy, I do not think that the obiter remarks of Lynskey J as to defective joists have any bearing on the present case.[85]

If it is necessary to go beyond Lister v Lane and Nesham, then I think that the decisions in Wright v Lawson (the bay window case) and Pembery v Lamdin (the damp basement case) are ample authority for the view that this doctrine of subsidiary parts does not throw on the lessee an obligation to provide an improvement to eliminate an inherent defect, though affecting only a part of the building; and I so hold. It follows that the tenant is entitled to succeed on the preliminary issue.[86]

Forbes J deals with these comments of Sir Brett Cloutman VC as follows:

In these passages it seems to me that Sir Brett Cloutman misdirects himself on the ratio of Sotheby's case. The question of whether the inherent nature of the building might result in its partial collapse was not obiter at all. It was part of the ratio in this sense that, treating the question as a matter of degree, a partial collapse, in the view of Lynskey J, would have been of a degree which brought it within the tenant's covenant to repair, whereas a total collapse would put it outside. As, therefore, it was not a matter of part only, but of putting in new foundations in the entire building, Lynskey J found it was not within the ambit of the covenant. Insofar as he appears to be misdirecting himself on the ratio of Sotheby, the persuasive authority of Sir Brett Cloutman's judgment in Collins must be considerably eroded.[87]

We have already seen that the value of Lynskey J’s judgement on the law of inherent defects must be doubted insofar as he relied on a case that was off-point. It is, in any event, a single judge judgement. But even if that problem is waived, Forbes J’s claim that Sir Brett Cloutman VC misdirects himself on the ratio of Southeby is not warranted. Southeby did not concern subsidiary parts but the defective foundations underpinning the entire house. Clearly, the remark about what would be the case if the facts were different is par excellence an obiter remark, as indeed is indicated by the nature of Lynskey J’s own language in Southeby’s case: ‘It may be that the inherent nature of a building may result in its partial collapse. One can visualise the floor of a building collapsing, owing to defective joists having been put in.’[88]

As the expressions ‘it may be that’ and ‘one can visualise’ make clear, the comments of Lynskey J are obiter comments in so far as his Honour, by those comments, is speculating what would be the case if the facts were other than they indeed were in the case before him. True it is that Lynskey J treats the question as a matter of degree, but it does not follow that speculative comments as to which side of the line certain facts might fall are not obiter comments. Consequently, it is submitted that Forbes J’s criticisms of Sir Brett Cloutman VC’s commentary on Southeby must be rejected.

It follows, in my judgement, that Forbes J’s conclusions in Ravenseft cannot be said to be entirely consistent with the authorities. Not only does he make criticisms of earlier cases which are, in fact, unfounded, but he also relies on cases, such as Lurcott, which were off point in so far as the case did not concern inherent defects at all. In so far as Forbes J relies on that case for the test of whether the work required in Ravenseft falls within the repair covenant, his judgement must be considered questionable. Notwithstanding that the position in the UK is clear,[89] the rationale for that position – based as it is on arguably erroneous interpretation and application of previous cases – must be doubted. This, in particular, is a significant reason why Australia should not automatically follow the UK authorities on this issue.

V OBJECTIONS TO THE INTERPRETATION OF THE LAW ON INHERENT DEFECTS ADVANCED IN THIS PAPER

One problem with the approach recommended here concerns the applicability of the doctrine to substantial parts of a building. If the test concerning inherent defects differs, as I contend, from that concerning repair and renewal – so that, on my account, it would not be the case that the tenant is liable for the defect even where its removal does not require renewal of the whole, or substantially the whole, of the building – then wouldn’t this position make it extremely difficult to distinguish repair from renewal? We have seen above that the distinction between repair and renewal only acquires a grip when these terms are predicated, not of the parts of the building, but of the whole or substantially the whole. This is because, as we have seen, repair always involves some renewal. Recall the examples given by Buckley LJ in Lurcott:

A skylight leaks; repair is effected by hacking out the putties, putting in new ones, and renewing the paint. A roof falls out of repair; the necessary work is to replace the decayed timbers by sound wood; to substitute sound tiles or slates for those which are cracked, broken, or missing; to make good the flashings, and the like.[90]

The objection to my argument would be: how are we to police an intermediary test of whether, for example, the leaking skylight was caused by putties that were defective? What if, in investigating the cause of the disrepair to the roof, one discovers that the decayed timbers were caused by the builder’s omission to put the necessary proofing on the wood or a failure to use the right wood? Or what if it is subsequently discovered that one kind of timber is more durable than another kind of timber, even though, at the time the house was built, this was not known? Doesn’t this make the case on a par with the facts of Ravenseft?

In response to this criticism, it should be noted that taking into account an inherent defect, even as applied to parts, and not the whole, of the building, does not threaten the distinction between repair and renewal. We can distinguish cases where the putties need to be replaced because the original work was performed in a defective way, from cases where the putties need to be replaced simply because of the passage of time and natural wear and tear. Likewise, we can distinguish between tiles that crack prematurely – because the wrong ones were used, or because they were placed badly – and tiles that crack because of the passage to time, or because cheap tiles were permissibly used in the construction. Consequently, it is possible to decipher cases where the damage is caused by an inherent defect and cases where the damage is not so caused. There may be cases where the distinction is blurry – where cases will fall at the borderline – but this is not a particularly unusual problem in our law. At this point it will of course be a question of degree whether the problem results from an inherent defect or not, and each case would have to be decided on its own facts.

A second, perhaps more searching, criticism is this: if the defect concerns only a part of the building, and not the whole/substantially the whole, then there will obviously be cases where it would be wrong to say that the removal of an inherent defect makes the building one which is different in kind from that was originally handed over at the beginning of the tenancy. Suppose, for instance, that the work required concerned the removal and replacement of defective putty in the window frames. It would be ludicrous to hold that, once those putties are replaced, one is giving the landlord back a different building from that which was originally demised under the tenancy.

There are two responses to this criticism. First, one should distinguish between insubstantial and substantial inherent defects. It will be useful, at this point, to recall the words of Jordan CJ in Graham v The Markets Hotel Pty Ltd:[91] ‘if, when it was taken over, it was subject to some inherent defect of a substantial kind, an agreement merely to repair does not impose an obligation to remove the defect, but only to maintain the structure subject to the defect so far as this can be effected by repair.’[92]

It is only if the removal relates to an inherent defect of a substantial kind that the building will, if the defect is removed, be of a different nature from that originally demised. And whether or not the defect amounts to a substantial defect will be a matter of degree.[93] This response might, however, be seen to give rise to a fresh difficulty: does this not return us to the UK test after all? Wouldn’t removal of an inherent defect of a substantial kind mean that the building itself is being renewed? If that were so, isn’t the UK position, after all, a cogent one?

The objection that this response might give rise to a fresh difficulty merely assumes, however, that the test for whether an inherent defect is substantial, and the test for whether the work required to remove the defect requires repair or renewal, are the same test. But there is no reason to make such an assumption. One can conceivably have a defect of a substantial kind, which nevertheless does not require renewal of the whole building. Clearly, there is a difference between a building that is not fitted with expansion joints – and the consequences that ensue from that omission – and a building which has defective putties fitted around its window frames. Certainly the difference is one of degree, but there is nevertheless a significant difference between these cases. Arguably, in Ravenseft, a building with the necessary expansion joints and cladding was a building different in kind from one without the same, even though the work required was not held to amount to renewal, because the consequences of the defect were substantial. It is this consideration of a defect of a substantial kind that lies behind the decision of Sir Brett Cloutman VC in Collins v Flynn. There, and in the two important cases concerning inherent defects that Sir Brett Cloutman reviewed – Wright v Lawson and Pembery v Lamdin – the work required did not amount to the renewal of the whole of the building, but only of a subsidiary part. Nevertheless, because the inherent defect was of a substantial kind, Sir Brett Cloutman, following those decisions, held that it was outside the repair covenant. This shows that whether or not an inherent defect is of a substantial kind is decided according to a different test from whether or not the work required to remedy damage amounts to renewal of substantially the whole of the building. The two tests need not, and in fact do not, coincide. Yet, as we have seen, the test for whether the defect is of a substantial kind has, by some judges, most notably Forbes J in Ravenseft, been erroneously merged with, or, more accurately, effaced by, the test of whether the work required amounts to renewal of the whole.

A second, alternative response to the criticism is this. Unlike in the case of the distinction between repair and renewal, there is no reason to restrict the application of the distinction between inherent defect and disrepair to the whole or substantially the whole of the building. Unlike the former distinction, the latter can readily find application to parts of a building.[94] Consequently, the test of whether the tenant would be giving back to the landlord a different thing from that demised can be applied to the parts of the building just as much as to the whole. If so, then the tenant would not be liable, under the tenancy, to replace even the defective putties of a window frame – if the problem does indeed arise from, eg, defective workmanship or the use of the wrong materials, etc. This appears to be the position in Australia. In Graham v Markets Hotel Pty Ltd,[95] the inherent defect considered by Jordan CJ concerned a possible inherent defect in a lavatory, not an inherent defect in the entire structure of the building. Although, in the High Court, the case was decided on other grounds – it being held that there was no such inherent defect, Latham CJ endorsed Jordan CJ’s comments that, if there had been such an inherent defect, then the tenant’s obligation to repair would have been accordingly diminished.[96] These being obiter comments, the issue of whether the distinction between inherent defects and disrepair can be applied to parts, rather than the whole, of the building must be considered to remain unclear. But in response to the objection outlined above, it is only necessary to show that there is no reason in principle why the distinction would not be applicable to parts, not just the whole.

VI CONCLUSION

The Australian position as to the whether inherent defects and their consequences fall within the repair covenant is superior to the present English position. This is so for several reasons. First, removal of an inherent defect causing damage is not something that would reasonably be contemplated by the tenant under a repair covenant, because the tenant has no control over matters of construction which take place prior to the existence of the lease. Although there are obviously other matters over which the tenant has no control but of which he or nevertheless must assume the risk, those matters are usually risks that arise during the term of the tenancy, and to that extent the control that the tenant has over those risks is different in kind. They are the kinds of risk that are normally and reasonably contemplated by the parties when entering into the tenancy. Second, it seems somewhat peculiar that liability for an inherent defect causing damage should be made to depend entirely on the extent of the work required to remove it. Third, it is illogical that an inherent defect, not being damage but being a fault extant from the very beginning, does not of itself fall within the repair covenant, but does so if that defect causes damage. It is more rational that the defect, and the damage it causes, fall outside the repair covenant. Fourth, the UK authorities exhibit various anomalies and inconsistencies. In particular, we have seen that subsequent cases seem to have misconstrued the leading authority on the issue, Lister and Lane v Nesham.[97] Furthermore, the influential judgement of Forbes J in the case of Ravenseft is open to a number of objections, which seem to be to be unassailable. These objections must be seen to place question marks over the case’s authority, or at least its persuasive force in Australia. It is submitted that the position in Australia should remain that adopted in Lazar, Clowes and Graham v The Markets Hotel.

In the meantime, given the state of uncertainty concerning whether, if damage is caused by an inherent defect, a tenant will be liable under the covenant to repair, it is advisable that a prospective tenant of a stand alone commercial property seek some protection by way of a building survey which would reveal the presence of any inherent defects. In addition, it would be prudent to insert carve outs in the repair covenant to make it clear that the repair covenant does not place an obligation on the tenant to put right an inherent defect in the building and to put right any damage that is caused by the presence of an inherent defect.[98]

---

[*] LLB (Hons) (QUT), LLM (QUT), PhD (Essex), Assistant Parliamentary Counsel, Office of the Queensland Parliamentary Counsel, Brisbane.
[1] Surprisingly, however, there have been very few articles written on the topic. The most significant article, that of A Hood, ‘The Extent of the Modern Covenant to Repair in Commercial Leases’ (1997) 5 Australian Property Law Journal 53, 57-8, is referred to below. A Stanfield’s article, ‘The Shifting Foundations Underlying Repair Covenants & how they Affect the Landlord, the Tenant and Third Parties’ (1996) 17 Queensland Lawyer 31 provides a useful summary of the law in the area, but concentrates primarily on the effects on the law of the principles of Bryan v Maloney (1995) 128 ALR 163. Her other article, ‘A Landlord's Liability to Repair: when does it Arise and How far Does it Extend? (1995) 3 Australian Property Law Journal 209 also concentrates primarily on the relationship between the repair covenant and personal injury.
[2] Occasionally, the bearer of the obligation is the landlord. Usually, it is the tenant.
[3] The focus here is exclusively on the common law. I do not examine the position that obtains where certain leases, for example, retail shop leases, are governed by applicable legislation.
[4] These terms are not synonymous: Lurcott v Wakeley and Wheeler [1911] 1 KB 905, 916-7 (Fletcher Moulton LJ). Further, an obligation to put into repair is different from an obligation to maintain in repair: Lurcott v Wakeley and Wheeler [1911] 1 KB 905. For our purposes, it matters little whether the obligation is to put into repair, or whether it is to maintain in repair, etc: the issue is whether an inherent defect, and damage consequent thereon, can ever fall within the covenant to repair, regardless of whether the obligation is to put the premises into a state of repair when the lease is taken, or whether it is merely to maintain the premises in a state of repair.
[5] Lurcott v Wakeley and Wheeler [1911] 1 KB 905, 914: ‘Is what has happened of such a nature that it can fairly be said that the character of the subject-matter of the demise, or part of the demise, in question has changed. Is it something which goes to the whole, or substantially the whole, or is it simply an injury to a portion, a subsidiary portion...of the demised propety?’ (Cousens-Hardy MR). It should be noted that the first sentence of this quotation is slightly unhappy in so far as it does not sit so comfortably with the second sentence: the phrase ‘or part of the demise’ should, strictly speaking, be omitted, if the propositions enunciated by each sentence are to be made consistent. It should also be noted that three differing tests have been mooted. See McDougall v Easington District Council [1989] 1 EGLR 93, 96.
[6] Lister v Lane and Nesham [1893] 2 QB 212, 216-7 is the case most often cited for this proposition. As will be seen below, in my judgement, Lister v Lane has been misinterpreted by later judicial consideration of the point in relation to the question of inherent defects, about which, arguably, the case was exclusively concerned. Nevertheless, there is ample authority for the proposition: see Lurcott v Wakeley and Wheeler [1911] 1 KB 905; Brew Bros Ltd v Snax (Ross) Ltd [1970] 1 QB 612; Ravenseft Properties Ltd v Davstone (Holdings) Ltd [1980] 1 QB 12.
[7] Lurcott v Wakeley and Wheeler [1911] 1 KB 905.
[8] Italics added. Ibid 925.
[9] Ibid 905; Southeby v Grundy [1947] 2 All ER 761; Brew Bros Ltd v Snax (Ross) Ltd [1970] 1 QB 612; Ravenseft Properties Ltd v Davstone (Holdings) Ltd [1980] 1 QB 12; McDougall v Easington District Council [1989] 1 EGLR 93, 96.
[10] Antra Hood makes the following comments in relation to the application of the test: ‘Replacement of the expansion joints in a building (Ravenseft Properties Ltd v Davstone (Holdings) Ltd [1980] 1 QB 12), replacement of an entire roof (Elite Investments Ltd v TI Bainbridge Silencers Ltd [1986] 2 EGLR 43), replacement of old single glazed windows by double glazed UPVC windows (Sutton (Hastoe) Housing Association v Williams [1988] 1 EGLR 56), and reconstruction of a roof (with a value of 20% of the value of the building) (New England Properties v Portsmouth New Shops (1993) 67 P & CR 141) have been held to be repair. On the other hand, rebuilding a wall for £8000 (compared to the value of the repaired building, which was somewhere between £7500 and £9500) Brew Bros Ltd v Snax (Ross) Ltd [1970] 1 QB 612), replacement of the entire aluminium cladding of a modern high quality office building (Credit Suisse v Beegas Nominees [1994] EGLR 76, 89 (Lindsay J) and reconstruction of a low-cost house to make it watertight (McDougall v Easington District Council [1989] 1 EGLR 93), go beyond repair’, ‘The Extent of the Modern Covenant to Repair in Commercial Leases’, above n 1.
[11] Hood, ‘The Extent of the Modern Covenant to Repair in Commercial Leases’, above n 1, 56.
[12] [1989] 1 EGLR 93, 96.
[13] See Hood, ‘The Extent of the Modern Covenant to Repair in Commercial Leases’, above n 1.
[14] Graham and Anor v The Markets Hotel Pty Ltd [1942] NSWStRp 43; (1942) 43 SR (NSW) 98, 103.
[15] Hoffmann J in Post Office v Aquarius Properties Ltd [1985] 2 EGLR 105, 107, attributing the dictum to Sachs LJ in Brew Bros Ltd v Snax (Ross) Ltd [1970] 1 QB 612. My own search of Sachs LJ’s judgement in Brew has been unable to locate any dictum to that effect. See however, the comments of Lawton LJ in Quick v Taff-Ely Borough Council [1985] EWCA Civ 1; [1986] 1 QB 809, 821, which are reproduced in the following note.
[16] In Quick v Taff-Ely Borough Council [1985] EWCA Civ 1; [1986] 1 QB 809, 821, Lawton LJ puts it as follows: ‘as a matter of the ordinary usage of English, that which requires repair is in a condition worse than it was at an earlier time’. This cannot be said of a building that was defectively constructed.
[17] Ibid 817.
[18] Ibid; Ravenseft Properties Ltd v Davstone (Holdings) Ltd [1980] QB 12.
[19] Quick v Taff-Ely Borough Council [1985] EWCA Civ 1; [1986] 1 QB 809.
[20] Ibid 818.
[21] Ibid.
[22] [1987] 1 All ER 1055.
[23] Ibid 1063 (Ralph Gibson LJ) and 1065 (Slade LJ).
[24] In Post Office v Aquarius Properties Ltd [1987] 1 All ER 1055, 1063, Ralph Gibson LJ states that ‘the reasoning of the court in Quick’s case is equally applicable whether the original defect resulted from error in design, or in workmanship, or from deliberate parsimony or any other cause.’
[25] [1980] 1 QB 12, 21.
[26] Ravenseft Properties Ltd v Davstone (Holdings) Ltd [1980] 1 QB 12, 21. Ravenseft is discussed in detail in section 4 below.
[27] [1989] 1 EGLR 93, 96.
[28] Ravenseft Properties Ltd v Davstone (Holdings) Ltd [1980] 1 QB 12, 21.
[29] (1886) 7 NSWLR 98.
[30] Ibid 107 (Sir Martin CJ).
[31] Ibid 110 (Sir G Innes J).
[32] Ibid 107-8.
[33] Ibid 108-9.
[34] Ibid 110.
[35] [1942] NSWStRp 43; (1942) 43 SR (NSW) 98.
[36] Ibid 103.
[37] Latham CJ states: ‘as his Honour said, if, when the building was taken over, it contained an inherent defect of a substantial kind, the covenant merely to repair does not impose an obligation to remove the defect...’ [italics added] at 580. He then goes on to say that, on the facts, there was no such inherent defect, at 581.
[38] [1943] HCA 8; (1943) 67 CLR 567, 579.
[39] [1942] NSWStRp 43; (1942) 43 SR (NSW) 98, 103.
[40] [1970] WAR 24.
[41] Italics added. Ibid 27.
[42] [1942] NSWStRp 43; (1942) 43 SR (NSW) 98; (1943) 67 CLR 567.
[43] This may, however, be thought to leave open the possibility that the tenant would be under an obligation to remove it if there was only insubstantial damage, that is, damage that does not require work to prevent a collapse or the destruction of the premises. Furthermore, the case may be reconcilable with the English authorities in so far as it may be that his Honour is simply pointing out that an inherent defect is not itself damage, and so does not fall within the repair covenant until damage is caused. For this reason, it cannot be conclusively determined whether Clowes really does depart from the UK authorities on this point.
[44] (Unreported, Supreme Court of Tasmania, Wright J, 21 October 1986).
[45] [1911] 1 KB 905.
[46] [1970] 1 QB 612.
[47] [1893] 2 QB 212.
[48] It is noteworthy that in New Zealand, the UK position has been adopted. In Weatherhead v Deka New Zealand Ltd (No 2) [1999] 1 NZLR 453, a restaurant was found to be structurally unsound and notices were issued by the council requiring remedial work to be undertaken. The issue was whether the tenant was liable under the repair covenant. Having reviewed the authorities – but notably, without making reference in his judgement to Lister v Lane and Nesham [1893] 2 QB 212, 465 – Baragwanath J held that the work fell outside that required under the repair covenant on the ground that it constituted renewal. His Honour relied, in particular, on Lurcott v Wakeley and Wheeler [1911] 1 KB 905, and Brew Bros Ltd v Snax (Ross) Ltd [1970] 1 QB 612 even though those cases, as already noted, did not concern inherent defects. The UK authorities were also briefly mentioned in Flora Investments Ltd v Samson Corporation Ltd (1999) ANZ ConvR 399. There Paterson J, relying on Ravenseft Properties Ltd v Davstone (Holdings) Ltd [1980] 1 QB 12, 402 held that it is a matter of degree whether the work required can properly be described as repair, or whether, on the contrary, it extends beyond repair to renewal of the whole of the demised subject matter. As argued below, there are strong grounds for doubting the correctness of Ravenseft, which, if cogent, would have implications for the reliability of those cases which found their analysis on Ravenseft.
[49] [1989] 1 EGLR 93, 96.
[50] The contemplation of the parties test is used in Lurcott v Wakeley and Wheeler [1911] 1 KB 905; Ladbroke Hotels Ltd v Sandhu and Singh 72 P & CR 498. The UK authorities do not adequately explain the relationship between the renewal test and the contemplation of the parties test – partly, no doubt, because, as I have noted, the test for an inherent defect seems to have been effaced by the test for renewal. But it is not obvious that the result of applying the renewal test and the result of applying the contemplation of the parties test will always coincide. As argued here, the case of inherent defects would be a case in which it would not be contemplated by the tenant that he or she should have responsibility for it, even though its removal might not require renewal of the whole or substantially the whole of the building. It is noteworthy that the application of the contemplation of the parties test in Australia has yielded different results, precisely in the context of inherent defects. Thus in Lazar v Williamson and Others (1886) 7 NSWLR 98, Sir J Martin CJ said that the parties ‘did not in any way have in contemplation the expending of any money for making good any structural defects in the theatre’ at 107-8, and Faucett J said ‘they would not be liable if, by any structural defect in the building, the repairs needed are not such as were contemplated by the parties who entered the contract’ at 108-9. Note too, that in Graham and Anor v Markets Hotel Pty Ltd [1942] NSWStRp 43; (1942) 43 SR (NSW) 98, 103, Jordan CJ defined an inherent defect as a defect ‘of an abnormal kind’ which implies that it would not be something that might reasonably be contemplated by the parties.
[51] See for example the case of Martin v Croft (Unreported, Supreme Court of Tasmania, Wright J, 21 October 1986) discussed above.
[52] [1980] 1 QB 12, hereafter ‘Ravenseft’.
[53] Post Office v Aquarius Properties Ltd [1987] 1 All ER 1055, 1061 (Ralph Gibson LJ).
[54] Ravenseft Properties Ltd v Davstone (Holdings) Ltd [1980] 1 QB 12, 21.
[55] Ibid.
[56] [1893] 2 QB 212, 216-7.
[57] Other than in Ravenseft, the passage is mentioned in Lurcott v Wakeley and Wheeler [1911] 1 KB, 924; Brew Bros Ltd v Snax (Ross) Ltd [1970] 1 QB 612, 640 (Sachs LJ) and 645-6 (Phillimore LJ); Pembery v Lamdin [1940] 2 All ER 434 where Slesser LJ identifies the passage and says that effectively the same thing is said in Lurcott v Wakeley and Wheeler [1911] 1 KB – in other words, he takes Lord Esher’s passage in Lister to the test for renewal; McDougall v Easington District Council [1989] 1 EGLR 93, 95, where Mustill LJ refers to Lister along with Lurcott and Brew, among others, and argues that three separate tests to determine the extent of liability under the repair covenant exist but that, at least on the facts before him, the result of applying those tests was the same. It is to be noted that the three tests mentioned, discussed above, are formulations of the test for whether the work required amounts to repair or renewal. In McDougall and in Quick v Taff-Ely Borough Council [1985] EWCA Civ 1; [1986] 1 QB 809, in Post Office v Aquarius Properties Ltd [1987] 1 All ER 1055, and in the New Zealand cases of Weatherhead v Deka New Zealand Ltd (No 2) [1999] 1 NZLR 453 and Flora Investments Ltd v Samson Corporation Ltd (1999) ANZ ConvR 399, the correctness of Ravenseft’s construal of Lister has not been doubted. On the contrary, the judgement is referred to with unequivocal approval.
[58] Recall that an inherent defect is not damage: it may therefore exist notwithstanding the absence of damage in the building.
[59] Inevitably there will be borderline cases and whether a different thing is given back will be a matter of degree. Consider a building with defective putties fitted round the window frames. When the defective putties are removed and replaced with new ones, it cannot be said that the tenant is giving the landlord a different thing from that which was demised under the tenancy agreement. The defect would have to have substantial implications for it to be meaningfully said that a different thing is being returned. But the test for whether the defect is substantial (ie a different thing is being returned) need not coincide with the test for renewal. For the reasons why this is so, see the discussion below in section 5.
[60] Lister v Lane and Nesham [1893] 2 QB 212, 216.
[61] Ibid.
[62] Admittedly, this interpretation of the additional comment of Lord Esher MR might be thought to sit uncomfortably with his use of the word ‘repair’ in the passage. For on my interpretation, he should simply be stating that one not only considers the extent of the work required (in order then to see whether that work amounts to ‘repair’ or ‘renewal’) but also whether the covenant has been broken. But I think his use of the word ‘repair’ here can simply be explained by the fact that he has inadvertently made use of the ordinary concept of repair, rather than the technical, legal concept.
[63] [1947] 2 All ER 761.
[64] Ibid.
[65] [1924] 1 KB 716.
[66] Ibid 728 (Bankes LJ); 730-1 (Scrutton LJ); 733-4 (Atkin LJ).
[67] [1978] EWHC QB 1; [1979] 1 All ER 929, 935.
[68] [1963] 2 All ER 1068.
[69] [1911] 1 KB 905.
[70] [1963] 2 All ER 1068, 1072.
[71] [1940] 2 All ER 434.
[72] (1903) 68 JP 34.
[73] Ibid.
[74] [1963] 2 All ER 1068, 1071.
[75] [1911] 1 KB 905.
[76] (1903) 68 JP 34.
[77] [1911] 1 KB 905.
[78] [1940] 2 All ER 434.
[79] The evidence established that the work ‘necessitates removing all the panelling and battening from the walls, cleaning them down, raking out the brickwork, asphalting the walls, and building a 41/2 ins. wall inside to keep the asphalt in position, and laying new concrete floor about 4 ins. thick to prevent water coming under the walls, as it does at present, and trickling across the existing floor. With regard to the dampness along the flank wall of the house that is the corridor and side of the staircase, in my opinion the cause is the same penetration of water from the earth under the pavement, and in addition there is a patch of broken defective rendering to the cement plinth along the side of the house. Here it would be possible for the wall to be waterproofed externally’. This was the evidence of the surveyor: Pembery v Lamdin [1940] 2 All ER 434, 437.
[80] [1911] 1 KB 905.
[81] [1963] 2 All ER 1068, 1073.
[82] Ibid.
[83] [1947] 2 All ER 761, 761-2.
[84] [1963] 2 All ER 1068, 1073.
[85] Ibid 1074.
[86] Ibid.
[87] [1978] EWHC QB 1; [1979] 1 All ER 929, 936.
[88] [1947] 2 All ER 761, 761.
[89] The case has been followed, or approved, in Quick v Taff-Ely Borough Council [1985] EWCA Civ 1; [1986] 1 QB 809; McDougall v Easington District Council [1989] 1 EGLR 93; and Elite Investments Ltd v T I Bainbridge Silencers Ltd [1986] 2 EGLR 43.
[90] Lurcott v Wakeley and Wheeler [1911] 1 KB 905, 925.
[91] [1942] NSWStRp 43; (1942) 43 SR (NSW) 98, 103.
[92] Ibid.
[93] This is exactly the position taken by Jordan CJ in Graham v The Markets Hotel Pty Ltd [1942] NSWStRp 43; (1942) 43 SR (NSW) 98, 103, where he says: ‘The question whether a particular defect is inherent or matter of repair is one of fact and degree’, citing Lyon v Greenhow (1892) 8 TLR 457. Note that this statement differs from the statement that whether the work required is repair or renewal is a matter of degree. Here I am concerned with whether a defect is an inherent defect or not – a question that is completely irrelevant to the issue of repair versus renewal, on the UK authorities.
[94] It is, however, more likely that the distinction would be considered to apply to substantial parts, eg, the whole of one of the walls of the building, rather than any part, such as a window-frame.
[95] [1942] NSWStRp 43; (1942) 43 SR (NSW) 98, 103.
[96] Graham v Markets Hotel Pty Ltd [1943] HCA 8; (1943) 67 CLR 567, 581 (Latham CJ); 586 (Starke J).
[97] [1893] 2 QB 212.
[98] More detailed suggestions of an alternative repair covenant are given by Antra Hood in section 7 of her article ‘The Extent of the Modern Covenant to Repair in Commercial Leases’, above n 1. However, the clause suggested by Antra Hood would also need to include a carve out explicitly excluding from the repair covenant any obligation to put right damage which is caused by an inherent defect, and not merely the inherent defect itself.