This is the final instalment in my series of 12 posts on variations that has run (approximately) every other month for the last two years. I have tried to cover all the key issues that arise, such as whether a variation instruction must be in writing (and how to claim if it isn’t), whether an employer can be under a positive duty to vary the scope and whether a contractor can refuse to carry out an instructed change.
As many of these posts will hopefully have illustrated, while variations are the bread and butter of contractual disputes on a project, it is a subject that raises some fascinating issues that do not always get the attention they deserve.
Possibly the most overlooked issue in this area is the question of how construction contracts deal with delay caused by variations.
EOTs and LDs: variations are different
If an employer delays the contractor and the contract does not contain an effective extension of time (EOT) mechanism to extend the completion date, then the contractor will be entitled to a reasonable period of time to complete. This, in turn, will mean that there is no fixed completion date and liquidated damages (LDs) will no longer be enforceable because, it is said, time is at large.
It is sometimes suggested that this principle applies in the same way to all delays that the employer is responsible for. But this is not the case. If the employer’s breach of contract causes delay then this will be treated differently to situations where the employer has varied the works. This distinction has important consequences.
Employer breach causing delay
An employer has an implied duty of cooperation, which is also sometimes described as the prevention principle. One consequence is that the employer cannot, in breach of its contractual obligations, prevent the contractor undertaking its work and then take advantage of this by imposing liquidated damages.
If an employer’s breach delays completion of the project, the contractor is entitled to an additional period of time, assessed on a reasonable basis. In order to avoid this implied entitlement to additional time being triggered (and therefore time being at large), contracts contain express provisions to deal with the consequence, that is an extension of time mechanism that extends time in the event of an employer breach.
There is a long line of cases in which the employer’s breach, under a contract with no extension of time mechanism, resulted in it being unable to deduct liquidated damages. It goes back to the 1838 case of Holme v Guppy (1838) 3 M&W 387, where the employer prevented access. Other regularly quoted cases include Wells v Army and Navy Co-op Society (1902) 86 LT 764, where the employer failed to provide design information, and Peak Construction (Liverpool) Ltd v McKinney Foundations Ltd  1 BLR 111 where the employer did not approve drawings.
Variation causing delay
When an employer operates a contractual variation mechanism that empowers it to instruct changes to the scope of works then it is, of course, not in breach. It is simply making use of its contractual rights.
It may still be the case that the contract does not contain an extension of time mechanism and so there is no contractual trigger to extend time for the instructed variation. But the instructed variation cannot be treated as an act of prevention because it is not a breach of contract.
In such situations, the courts have said that time is put at large. But since there is no breach, the underlying logic must be different. To appreciate why, it is worthwhile thinking about a contractor’s entitlement to extra time for variations in the context of the contractor’s entitlement to extra money.
Compensation for variations
Money and time are the two forms of compensation that a contractor will expect to get when the employer instructs a variation. The method by which the extra money and time for variations is assessed will normally be set out expressly in the contract.
We are used to the fact that the way in which the extra money for variations is determined varies from contract to contract. For example, such compensation may be assessed by reference to a schedule of rates, the contract sum build-up or may be cost reimbursable. But if the contract is silent as to how variations are priced then a right to a reasonable sum of money for the instructed change will be implied.
The same logic applies to the other form of compensation that is due for variations: time. If the contract does not contain an express provision for calculating the extra time due (that is, an extension of time mechanism), then the contractor will be entitled to a reasonable period of time. This is the logic that is used where two parties enter into a contract without agreeing a time period for the works – a reasonable period will be implied by the courts.
Because a reasonable period of time is due for an instructed variation (where there is no extension of time mechanism), then the completion date will not be re-fixed with certainty and, as a result, time will be at large. This has certainly been the approach the courts have traditionally adopted and leads to the same end result as an employer’s breach of contract.
Parties agree zero compensation for variations
A party’s implied right to extra money or extra time for a variation is, of course, only triggered if there are no express provisions in the contract to the contrary.
The contract will normally contain a valuation of variations clause. But the parties could agree that there is no compensation for variations. For example, there is no reason why the parties could not expressly agree that the first £200,000 worth of variations are undertaken for free.
Equally, the parties could agree that the contractor gets no extra time for instructed variations. In Jones v St John’s College (1870) LR 6 QB 115 such an arrangement was considered in detail. The contractor had been hired to build a farmhouse and, in the course of construction, the employer ordered additional work under the contract variations clause. The contract provided that the contractor was obliged to complete the scope, including undertaking any extra work ordered, by the original completion date. The court found that the provisions of the contract were clear and there was no basis for implying a right to additional time where none existed.
Under such a contract, the delay caused by a variation does not put time at large even though there is no right to an extension of time. This is because the parties are free to agree under their contract the compensation (that is, money and time) that the contractor is due when a variation is ordered. Therefore, the approach to variations is quite different to breaches of contract.
Re-fixing the completion date
In the context of the discussion above, it is valuable to re-analyse why time is put at large and how this operates for variations.
It has long been recognised that the instruction of a minor variation under a contract that does not expressly allow for an extension of time to be given can have unexpected and surprising consequences. This issue was the subject of a talk given by Ramsay J to the Society of Construction Law in April 2012, which referred to the following passage from Coleman J’s judgment in Balfour Beatty v Chestermount (1993) 62 BLR 1, which succinctly illustrates the point:
“The remarkable consequences of the application of this principle could therefore be that if, as in the present case, the contractor fell well behind the clock and overshot the completion date and was unlikely to achieve practical completion until far into the future, if the architect then gave an instruction for the most trivial variation, representing perhaps only a day’s extra work, the employer would thereby lose all right to liquidated damages for the entire period of culpable delay up to practical completion or, at best, on the respondents’ submission, the employer’s right to liquidated damages would be confined to the period up to the act of prevention. For the rest of the delay he would have to establish unliquidated damages. What might be a trivial variation instruction would on this argument destroy the whole liquidated damages regime for all subsequent purposes.”
It should be remembered that the reason why time becomes at large is that without an extension of time provision, there is supposedly no mechanism to re-fix the completion date and therefore no clearly fixed completion date from which to impose liquidated damages.
While this may make sense for breaches of contract, it is difficult to understand in circumstances in which a variation has delayed completion.
If a contract does not contain a clause setting out how a variation is priced then the contract administrator will still be able to assess the sum due. The processing of the next monthly valuation will not fall apart simply because the contract does not contain express provisions for the pricing of the variation.
The time at large logic also seems somewhat out-of-date in the context of adjudication. The cases on this subject all date from a time when a party would have to go through a two-year court process in order to have its completion date re-fixed following the instruction of a variation. But adjudication allows a third party to re-fix the completion date using a procedure that can take as little as 28 days. This would resolve the problem raised by Coleman J in Balfour Beatty v Chestermount, and would allow the re-fixing of the completion date for the purposes of deducting liquidated damages.
Delay caused by variations therefore raises particular issues. The distinction between breaches and variations causing delay is an important one that should not be blurred.