Recent decisions in England, Australia and New Zealand highlight the uncertainty surrounding the interpretation and application of exclusion clauses that use 'catch-all' language such as 'consequential' or 'indirect' loss.
This approach - as opposed to clauses that detail the types of losses that are recoverable or excluded - means that courts are left with considerable room to manoeuvre when deciding what claims to allow. In some cases this may lead to surprising outcomes.
In England, the terms 'consequential' or 'indirect' are equated with loss under the second limb of the rule in Hadley v Baxendale (1854) 9 Ex 341 and do not apply to loss recoverable under the first limb of the rule.
In Australia and New Zealand ('ANZ') several courts1 have moved away from the English position and worked on the basis that 'indirect' or 'consequential' loss means loss which is not 'normal loss' in (arguably) the sense used in earlier editions of McGregor on Damages. In those cases, 'consequential' or 'indirect' loss could be first or second limb loss under the rule in Hadley v Baxendale.
Moreover, the cases considered in this article suggest that the ANZ courts are likely to give a more exclusionary effect to broadly-drafted exclusions of consequential loss than their English counterparts.
That said, neither approach is entirely satisfactory and the better overall approach may be to avoid use of general or 'catch-all' terminology altogether in favour of more detailed clauses that list the types of loss that are recoverable. The preparation of a model 'checklist' endorsed by the IT industry may be worth considering.
We begin with a discussion of these issues by reference to a small selection of key cases in England and ANZ, highlighting the different approaches and some problems with each case. Although most are not IT cases, they are nonetheless relevant.
Croudace Construction Ltd v Cawoods Concrete Products Ltd [1978] 2 Lloyd's Rep 55 was decided by the English Court of Appeal in 1978 and concerned a contract for the supply of goods which were delivered late. The relevant exclusion clause provided as follows:
We are not under any circumstances to be liable for any consequential loss or damage caused or arising by reason of late supply or any fault, failure or defect in any material or goods supplied by us or by reason of the same not being of the quality or specification ordered or by any other matter whatsoever.
The Court, following Millar's Machinery Co Ltd v Way [1934] 40 Com Cas 204, held that the reference to consequential loss meant loss recoverable under the second limb of the rule in Hadley v Baxendale - i.e. loss that may reasonably be supposed to have been in the contemplation of the parties at the time of formation as the probable result of the breach (sometimes referred to as 'special loss').
Under this approach, exclusions of consequential loss do not apply to loss under the first limb of the rule in Hadley v Baxendale - i.e. loss arising naturally (or 'according to the usual course of things') from the breach.
The Croudace approach has been consistently applied over the past 35 years,2 including in relation to IT cases. In a sense, the term 'consequential' has become a term of art. For completeness, it should be noted that the UK Supreme Court has yet to confirm that the Croudace approach is good law. Indeed in Caledonia North Sea Ltd v British Telecommunications PLC [2002] UKHL 4, Lord Hoffman expressly reserved for another time the question of whether the approach was correct.
The decision by the English Court of Appeal decision in GB Gas Holdings Limited v Accenture (UK) Limited [2010] EWCA Civ 912 (GB Gas) provides a useful (and relatively recent) example of the application of the English approach. It is also a case in the IT area.
In GB Gas the Court of Appeal applied Hadley v Baxendale and found that the following losses (if proven to arise from breaches by Accenture of a contract to supply an automated billing system) fell within the first limb of the rule in Hadley v Baxendale and were therefore recoverable:
The Court made these findings even though the exclusion clause expressly sought to exclude liability for:
As we shall see, it is probable that the courts in ANZ would form a different and more exclusionary view on the facts of GB Gas.
First, it may be unrealistic to assume that ordinary business-people involved in negotiating contracts have Hadley v Baxendale in mind when agreeing exclusion clauses. How likely is it that they would think of consequential loss in terms of a technical legal rule from the 19th Century founded upon concepts of foreseeability? Even if ordinary business-people were assumed to have the legal expertise necessary to understand the rule, it is difficult to see how (absent particular contextual clues) they would ordinarily interpret the words 'consequential' or 'indirect' as referring to losses attributable to a special susceptibility of the plaintiff.
Indeed, if an ordinary businessperson heard about the decision in GB Gas, they might (with some justification) wonder what, if anything, could ever be caught by the exclusion of 'indirect or consequential' loss in that case.
Secondly, the rigidity of the English interpretive approach denies the possibility that parties in a given case could intend 'consequential loss' to mean first or second limb liability, depending on the particular circumstances involved. Why rule out the possibility that some consequential loss, when construed in the light of the particular wording used and wider context, could be loss that falls within the first limb of Hadley v Baxendale?
Thirdly, in the absence of detailed provisions specifying recoverable or excluded losses, it will remain very difficult for parties (and their advisors) to predict what the judicial outcome might be if and when individual clauses are litigated.
In Environmental Systems Pty Ltd v Peerless Holdings Pty Ltd (2008) 19 VR 358; [2008] VSCA 26 the Victorian Court of Appeal moved away from the English position described above. This case concerned the supply of a regenerative thermal oxidiser ('RTO') to replace a less efficient afterburner used to limit noxious emissions from an animal rendering plant. The RTO did not perform as expected and Peerless (the purchaser) was forced to reinstate the existing afterburner. The relevant exclusion clause provided as follows:
As a matter of policy, Environmental Systems does not accept liquidated damages or consequential loss. Environmental Systems is motivated to achieving agreed milestones through respect for the client's needs and the obvious financial advantage gained from completion of projects in the shortest possible period.
In this case, the Court interpreted 'consequential' by reference to the dichotomy between 'normal loss' and 'consequential loss' described in earlier editions of McGregor on Damages.3
The Court referred to normal loss as 'the loss that every plaintiff in a like situation will suffer', citing an extract of McGregor to the effect that in the context of goods sold or services rendered, normal loss will be the difference between the contract price and the market value of what (if anything) was supplied.
The following extract summarises the Court's conception of 'consequential loss' and its rationale for departing from the English approach (Nettle JA at [93]):
In my view, ordinary reasonable business persons would naturally conceive of 'consequential loss' in contract as everything beyond the normal measure of damages, such as profits lost or expenses incurred through breach. Despite the construction which has been put on 'consequential losses' by cases such as Millar and Croudace, it would be unrealistic to suppose that the appellant and the respondent employed the expression 'consequential loss' in cl 8.9 of the agreement advisedly in that sense. It is more likely in this context that they intended the expression to have its ordinary and natural meaning. Accordingly, I would construe the expression 'consequential loss' in cl 8.9 as intended to have that meaning. Read in the light of the contract as a whole, and giving due weight to the context in which the clause appears, including the nature and object of the contract, I see no ambiguity which as a matter of principle would warrant a departure from that view. It follows as I see it that, although the judge's approach in this case was in accordance with the English cases, it was not correct to construe 'consequential loss' as limited to the second rule in Hadley v Baxendale. [emphasis added]
On the facts, the Court concluded that:
These losses were therefore recoverable.
First, it may be unrealistic to attribute an understanding of the McGregor approach to ordinary business-people involved in negotiating IT contracts. This mirrors the criticism that the English decisions assume ordinary business-people understand Hadley v Baxendale.
Secondly, it is unclear whether the Victorian Court of Appeal intended that the governing formulation of normal loss should be the extract of McGregor quoted in the judgment, which referred to the difference between the contract price and the market value of what (if anything) was supplied, or the Court's paraphrased expression of that principle as the 'loss that every plaintiff in a like situation will suffer'.
Thirdly, if the paraphrased form is to apply, it seems at least conceivable that 'profits lost or expenses incurred through breach' (i.e. the examples of consequential loss cited by Nettle JA) could be normal loss depending on the particular circumstances involved.
Fourthly, it is possible that the Victorian Court of Appeal misapplied its own test (however formulated), when it found that items 1(b) (installing and commissioning RTO), 1(c) (attempting to make the RTO functional) and 1(d) (repairing the existing afterburner) were recoverable. On one view, each of these appears to be in the nature of 'expenses incurred through breach' (i.e. consequential loss, according to Nettle JA) and therefore excluded.
Moreover, the Court held that the value of the RTO as delivered was no less than the purchase price, suggesting that item 1(a) (purchase price of the RTO) should not have been recoverable, at least on the quoted formulation of 'normal loss'.
Finally, it should be noted that the approach in Environmental Systems has yet to be considered by the High Court of Australia (the equivalent to the Supreme Court in the UK).
Environmental Systems was highly influential in the recent Supreme Court of South Australia decision in Alstom Ltd v Yokogawa Australia Pty Ltd & Anor [2012] SASC 49. Alstom concerned the refurbishment of a power station in South Australia. A sub-contract between Alstom and a joint venture called YDRML contained the following exclusion clause:
Notwithstanding any other Article of the [subcontract], the Subcontractor shall not be liable for any indirect, economic or consequential loss whatsoever.
Alstom was able to claim liquidated damages for delay and reimbursement of upstream performance guarantee payments because they were expressly provided for under the contract. In addition however, Alstom made various other claims for compensation, including:
In considering these additional claims, the Court noted the exclusion clause and said it should be construed strictly against the person whose liability it purported to exclude. Nevertheless, having closely considered the context in which the clause appeared, Bleby J expressly referred to the reasoning in Environmental Systems described above and stated that he preferred that reasoning which, as a matter of precedent, he regarded as more persuasive than the English cases. On the facts, His Honour concluded that none of the additional claims for compensation listed above was recoverable.
Two other Australian courts5 have also indicated that the English position is not determinative of the interpretation of exclusions of 'consequential loss', however it was not necessary in either case for the court to reach a concluded interpretation of the term in the respective contexts.
In Oceania Furniture Ltd v Debonaire Products Ltd [2009] NZHC 1139, the High Court in Wellington said that it preferred the Australian approach over the English position.
In that case, a contract for the supply of furniture components contained a clause that sought to exclude liability for 'consequential, indirect or special damage or loss of any kind'.
Debonaire's business involved assembling furniture components supplied by Oceania and supplying the assembled furniture to retailers. Debonaire alleged breach of contract by Oceania(late or non-delivery of the furniture components) and made claims for:
Following the approach in Environmental Systems, items 1 and 3 were held to be consequential and therefore excluded. Item 2, on the other hand, was held to be recoverable - not just the difference between the purchase price of the components ordered and the value of the components actually supplied.
For completeness, it should also be noted that the ANZ judgments described above have, to some extent, attempted to look at the actual intention of the parties, by considering the context in which the relevant words appear. The ANZ approach described in this article should therefore be understood as a description of previous decisions, not a prescription for future ones.
For instance, it should be open to an ANZ court to interpret an exclusion clause in a manner similar to the English approach if the clause mirrors the language used in Hadley v Baxendale: where, for example, it goes further than the use of 'catch-all' terminology and defines 'direct loss' by reference to the wording used in the first limb of the rule in that case.
Despite the uncertainties arising from both the English and ANZ approaches, the broad message from the cases considered in this article appears to be that where 'catch-all' language such as 'indirect' or 'consequential' is used, the ANZ courts are likely to give a more exclusionary effect to exclusions of consequential loss than their English counterparts. To illustrate, the following table sets out how the ANZ courts might have treated the claims in GB Gas.
To recap: what we have seen in this article is that courts in different jurisdictions have adopted different approaches to the interpretation and application of clauses that seek to exclude liability for 'indirect' or 'consequential' loss. We have also seen that these terms give rise to uncertainty and may not deliver outcomes the parties were expecting. Plainly this represents an unsatisfactory state of affairs, particularly for those individuals charged with the responsibility of considering risk allocation and advising thereon.
In view of this, the question which arises is whether contracting parties should take matters into their own hands by adopting an approach that involves setting out, in detail, the actual types of loss to be recovered or excluded in each case.
We appreciate that achieving this will require buy-in from all concerned and that it may be challenging for parties to agree to undertake a detailed allocation of responsibility for individual heads of loss at the outset of a new relationship. The parties may, for example, be put off by being asked to consider the detailed consequences of a failed relationship before it even starts.
Moreover, a supplier operating in ANZ may prefer the use of general or 'catch-all' language on the basis that it stands a better chance of defeating future claims by customers - bearing in mind the recent approach of the ANZ courts discussed above. They may therefore try to resist any attempts by customers to include a detailed list of recoverable losses in contracts between them.
On the other hand, some suppliers may favour the inclusion of a list and the certainty this brings.
At least they will know where they stand and can look for concessions elsewhere in the contract - for example, in the area of pricing if the risk profile is higher through the inclusion of a detailed list.
We also appreciate that the idea of expressly specifying heads of loss in detail is not new and that some attempts at this have been made over the years. That said, none of the standard form consequential loss clauses considered in preparing this article goes beyond the use of high level or generic terms such as 'additional costs', 'wasted expenditure' and 'anticipated savings' - which provide only limited additional clarity about the heads of loss intended to be allowed.
We propose that consideration be given to the preparation of a detailed checklist of heads of loss relevant in IT contracts based on the substantial body of IT case law that has built up over the past 20-30 years. This list could then be used as a 'menu' (or starting point) by parties when negotiating exclusion clauses.
An endorsement by the IT industry, coupled with model clauses, should add considerable weight and maximise the prospects of the initiative achieving a measure of success in due course.
If it can be achieved, the potential advantages include:
The writers hope that this article provides a useful summary of the main issues in this area and welcome any feedback or comment that readers may have.
1 Australian cases referred to in this article can be obtained from www.austlii.edu.au and New Zealandcases from www.nzlii.org.
2 See e.g. British Sugar Plc v NEI Power Projects Ltd (1997) 87 BLR 42; Deepak Fertilisers and Petrochemicals Ltd v Davy McKee (London) Ltd [1999] 1 Lloyd's Rep 387; Hotel Services Ltd v Hilton International Hotels (UK) Ltd [2000] BLR 235; Watford Electronics Ltd v Sanderson CFL Ltd [2001] BLR 143.
3 The cited version was McGregor on Damages, 15thed, (1988) at [26] and following.
4 For a more detailed criticism of Environmental Systems, the writers recommend the learned article by Professor John Carter, Exclusion of Liability for Consequential Loss(2009) 25 Journal of Contract Law 118. Professor Carter discusses in some detail how the 'normal loss' approach (at least as adopted in Environmental Systems) is more similar to the Croudaceview than first appears, and is susceptible to many of the same criticisms.
5 The New South Wales Court of Appeal in Allianz v Waterbrook [2009] NSWCA 224 and the Western Australian Court of Appeal in SMEC Australia v Valentine Falls Estate Pty Ltd [2011] WASCA 138.