As a matter of contract, the pool's diving depth was to be 7.5 feet deep. Once completed, it was found that the depth was only 6 feet - deep enough, but not what was contracted for. Presumably those missing 1.5 feet saved the contractor some money, but it was significantly less than what the client wanted. The cost of rectification of the problem was significant - £21,560, in early 1990s worth. The owner had decided not to rebuild the pool, but wanted compensation. This was an admission fatal to the claim. In Ruxley Electronics v Forsyth  A.C. 344, it was held that what the claimant intended to do with its compensation is not normally its concern, but as the pool was still good for diving, there was no loss in value, and the cost of rectification out of all proportion to any benefit obtained, the claimant got nothing.
There is more than a little that is uncomfortable about this decision; not least the recognition that there are no hard and faster definitions to be adopted; but, like pornography, "you know it when you see it". What must have been galling for Mr Forsyth is that in an earlier, unreported judgment, the court of first instance held that he was not entitled to withhold payment for the pool, practical completion could be certified and that Mr Fosyth's sole remedy was in damages; damages which the House of Lords ultimately held he was not entitled to.
In Mears Ltd v Costplan Services (South East) Ltd  EWCA Civ 502, the Court of Appeal had the opportunity to review this issue in greater detail. The case concerned the construction of student accommodation for Mears. Under the agreement, Mears was to take over the lease of the new building on practical completion. It was a term of the agreement that there were to be no variations from the agreed plans which "materially affect the size" of the rooms by more than 3%. It was found in an earlier hearing that 56 of the rooms were smaller by more than 3%. Mears raised two arguments - first, that that the contract term drew a "red line" beyond which there was a material or substantial breach beyond which it was entitled to termination; and second, that the existence of a patent breach (one which the certifying professional was aware of) meant that practical completion should not have been certified.
The Court of Appeal said no on both counts.
On the first, it held that while material variations were prohibited ("material" being specifically set at greater than 3%), that did not make for a material breach. At first blush. this can only look like dancing on the head of a pin, but the point is expressly acknowledged that if the parties wished to record that this was a substantial or material breach entitling Mears to terminate they could have said so. They didn't.
For our own concept of substantially reducing the benefit or increasing the burden of a contract, in terms of section 37 of the Contract and Commercial Law Act 2017, a similar outcome has to be likely. Much like the dispute over the height of the window sills in the Rob't Jones Tower in Auckland in the 1990s, the courts are unlikely to grant such relief except in the most egregious of cases. More critically, it is for the parties to determine what is material and what is not, and in terms of both the case law and section 40 of the Contract and Commercial Law Act which allows the parties to provide for their own remedies in contract.
The second issue, that the owner's agent should not have issued the practical completion certificate, is more nuanced. The Court of Appeal went through six issues it considered relevant:
(1) Regrettably, practical completion is easier to recognise than to define.
(2) Latent defect, because they are by definition unknown, will not prevent practical completion.
(3) Known defects are either yet to be completed or are defective. In this connection, whether there is outstanding work or defects makes no difference, as snagging lists are common in the industry (for a somewhat contrary view, see the findings of Hinton J in Richina Pacific v AAI Ltd & Samson  NZHC 1686, at para  et seq).
(4) Fourth, practical completion generally means "free from patent defects, other than ones to be ignored as trifling" (see Mariner International Hotels v Atlas  10 HKCFAR 1).
(5) Fifth, whether or not a matter is "trifling" is a matter of degree, measured against whether or not the work can be taken over and used for its intended purpose. The converse is also true - the fact that the work can be taken over for beneficial use does not mean that it is practically complete.
(6) There is no authority (with the exception of the first instance decision in Ruxley referred to above) on the interplay between irremediable breach and practical completion. The mere fact that a defect cannot be remedied, it does not automatically follow that the work is not practically complete.
While we are left with the entire issue being a question of degree, the court has canvassed a number of issues on both points which do help to illustrate how this often difficult issue can be analysed. It is also certainly an object lesson for contract drafters to consider the issues of material breach giving rise to an entitlement to cancel, and the requirements for practical completion (and in particular any issues of irremediable breach), with care.