Re-examining Limitation of Liability Provisions

July 10, 2011

Given how much debate there is around limitation of liability clauses in outsourcing contracts, you would be forgiven for thinking there is nothing ‘new’ to discuss, and that instead all that is left is the rehearsal of ‘standard’ arguments as to the amount of the financial caps, etc. Surprisingly, however, there are a number of areas where a deeper examination of current market practice gives more than a little food for thought, and perhaps suggests that some of the ‘usual’ forms of drafting should be more critically evaluated. 

1.         Exclusion of Indirect Losses 

It is standard in outsourcing/services contract for there to be an exclusion of indirect loss. It is so usual to see this that rarely does anyone pause to question the rationale behind the exclusion. If an answer were sought, it would likely be one or other of the following: 

(a)    that such losses are too far removed, or too speculative, for it to be reasonable for them to be recoverable; or 

(b) (where such losses are claimed by a customer) that they have only arisen due to a set of special circumstances applicable to that customer, which will not have been factored into the price. 

One would think that both of these arguments would have to be robust to justify what amounts to almost a blanket ban on recovering indirect loss. But are they? 

As to the first argument, bear in mind that an injured party has to (i) prove the breach, (ii) prove the breach caused the loss in question, (iii) show that the loss is not too remote (ie that it must have been a reasonably foreseeable consequence of a breach, as at the date that the contract was signed), (iv) be able to accurately quantify the loss, (v) show that it has not itself been responsible for causing the breach and (vi) have taken steps to mitigate any loss suffered. Are we really saying, having gone through all of those barriers, that there should then be yet another test, which excludes a component of loss completely notwithstanding the fact that it would otherwise be legitimately recoverable?  If explained in this way, a lot of senior executive ‘lay’ clients might well question what their legal advisors were thinking! And yet that is what the automatic exclusion of indirect loss achieves.  

As to the second argument, there may be some justification for such a rationale when buying a standard software product or commodity style service offering. This is because in these contracts there is not usually a detailed analysis of the risks that the particular customer would face, and the price may be wholly ‘standard’ or based on rates with limited variability. But does this argument really stand up in relation to outsourcing contracts, where the parties have spent months (sometimes over a year) in detailed negotiations, where the full implications of failure are well understood, and the pricing will, in reality, have factored in the associated risks? 

If acting for a customer, we believe there is ample justification for indirect losses to be ‘re-habilitated’, and seen for what they are – losses which are not remote, which the parties know could occur and which in the larger contracts will (or should) have  been factored into the risk profile of the contract which has been used to set the price.  

Acting for the supplier, one would argue that the approach to the setting of the overall limit of liability was predicated on it applying only to ‘direct’ loss, that being the market norm at present. If the supplier is to run an ‘increased’ risk by reason of also accepting liability for indirect loss, it may as a result also require the setting of a lower overall liability cap.  This may however be a better overall result for both parties, given the increase in certainty it brings. 

2.         ‘Paid and Payable’ 

This phrase often appears in limitation of liability clauses. It is used generally as way of setting a financial limit instead of or along with a more straightforward fixed amount set as a ‘de minimis’ limit of liability. Hence, a paraphrase of a typical clause will read: 

Party A’s aggregate liability for any and all claims arising pursuant to or in connection with this Agreement is limited to the greater of £X or the amounts paid and payable under this Agreement. 

An obvious question, but one which does not often get discussed in sufficient detail, is what this phrase actually means. Though it may be thought to approximate the ‘value of the contract’, it is not the same. There appear to be at least four plausible interpretations, as a short example shows: 

Example – contract commences 1 Jan 2010 with £100K quarterly payments, due in arrears on end March, June, Sep and Dec. Duration 5 years, with ability of customer to terminate on 2nd anniversary or any quarter end thereafter.

Contract is terminated by supplier for customer breach on 1 Feb 2011 (in other words, one month into second year).
At this point three quarterly payments (totalling £300K) have been paid, one quarterly payment (of £100K for period to end Dec 2010) has been invoiced but not paid, and sums have accrued in respect of charges for the services provided during Jan (33K), but have not as at the point of termination been invoiced.

Supplier sues for £1 million based on losses suffered. What is the cap on liability?

(a)       £400K (£300K paid, with a further £100K ‘payable’ in respect of invoiced charges)

(b)       £433K (£300K paid, with a further £133K ‘payable’ representing charges invoiced but unpaid and also charges accrued but not invoiced)

(c)       £800K (£300K paid with a further £500K ‘payable’ until the earliest end date of contract (2 years)

(d)       £2 million (£300K paid with a further £1.7 million ‘payable’ for the full lifetime of the contract). 

As can be seen, in this example, the highest potential interpretation of the limit of liability is a sum amounting to 500% of the lowest potential amount – quite a difference! The solution seems simple – ‘paid and payable’ requires a definition. In the absence of one being set by the courts, we would suggest either (b) or (c) above (depending on whether acting for supplier or customer).  

3.         Calculation of the Cap 

What ‘counts’ towards the calculation of the cap? 

In most outsourcing contracts, there is unlimited liability for IPR infringement in some shape or form. Therefore, whether the liability cap is otherwise set at £1 million or £100 million, a claim for IPR infringement has no bearing on, nor relationship with, the cap on liability. Right? 

Consider, however, this scenario: 

(a)       An outsourcing contract has been signed. The following is an extract from the limitation of liability clause:

20.1 Subject to Clause 20.2, the liability of each party under this contract is limited, in the aggregate, to £30 million.

20.2 Nothing in this Agreement shall limit or exclude any claims arising under:

20.2.1  the indemnity under Clause 15 (IPR Indemnity). 

(b)       In year 2, a claim under the IPR indemnity is made and settled for £25 million. 

(c)       In year 5, the supplier is in material and repeated breach. The customer wishes to terminate the contract. The customer estimates its loss as being £15 million. Looking at Clause 20.1, the customer hopes that it can recover all of that amount, as it is within the aggregate cap, however the customer takes legal advice to confirm the position. 

(d)       The customer is surprised to be advised that while the IPR claim was not subject to the cap, the cap was nonetheless affected by the IPR claim, in as much as £25 million of it was ‘used up’ by the claim being settled. Therefore, the customer’s maximum amount recoverable exposure for material breach leading to termination is £5 million. 

The solution here is again relatively simple – ensure that the contract has the following proviso, in addition to the wording above in Clause 1.2:

No amounts awarded or agreed to be paid under Clause 15 (IPR Indemnity) shall count towards the cap on liability as set out in Clause 20.1

4.         Annual Caps on Liability  

Often issues can arise due to the drafting of so-called ‘annual’ caps.  Do we know what they actually mean, in practice? 

Take this example: 

The Supplier’s liability arising pursuant to this Agreement shall be limited to an annual cap of £1 million. 

So, does the ‘annual’ part of this refer to when claims were made, or when the relevant causes of action arose?  Are the years in question calendar years or consecutive years following the execution of the contract?  And what if linked causes of actions or claims span more than one year – do multiples of the limit then apply? Care should also be taken where annual caps are set out by reference to claims during each year of the contract – for what then is the position for claims made after the contract is terminated? 

Again, the solution is to be more explicit in the form of wording being used. We tend to err in favour of maintaining an aggregate limit of liability which applies to all claims arising at any point during the term of the contract but which flexes by reference to an agreed percentage of the average charges across the elapsed period of the contract to date (and perhaps with the addition of a ‘de minimis’ amount and/or an assumed level of charges for the first 12 months, until the actual level of charges across that period is known). 

5.         Loss of Profits 

It has long been ‘market standard’ for service providers to exclude all liability for any element of the customer’s loss which relates to loss of profit/revenue.  Historically this was due to a somewhat misconceived assumption that such loss would in any event be indirect. But, having had the courts quash that theory in British Sugar, service providers remained reluctant to take on this risk and so have tended to modify limitation of liability clauses to exclude loss of profit, regardless of whether it might be classified as either direct or indirect loss. 

There are some sound commercial grounds for this, not least when service providers see customers pressing for higher and higher caps, and their overall risk increasing.  They can also legitimately point out that their actual margins are but a percentage of the overall charges, and so accepting liability for the potentially significant loss of customer profits (as opposed to the costs of sourcing an alternative supply of services) would constitute a disproportionate allocation of risk. 

Whilst this may hold true in many, if not the majority of, cases involving outsourcing projects, there are good grounds for questioning whether it should always be so.  If for example the outsourcing services include the support of services and/or network components which underpin transactional elements of the customer’s business (such as trading web sites), it can readily been seen that defaults by the service provider would be highly likely to result in losses of revenue (and therefore profits); having to then explain to senior executives why none of this loss is recoverable from the culpable service provider may be a difficult conversation! 

In recent negotiations, therefore, we have seen the preconceptions regarding exclusions of loss of profit being challenged.  Alternative outcomes have included (a) setting some additional ‘super service credits’ in lieu of an ability to claim loss of profit, but payable for defaults likely to cause such loss; (b) sub-caps within the overall limit of liability so as to allow a (smaller) proportion of loss of profit to be claimed; and (c) allowing loss of profit to be claimed, but in return for a lowering of the overall limits of liability.  

6.         ‘Deeming’ Losses to be Direct 

Coming back to the somewhat uncertain distinctions concerning what might be classified as direct/indirect loss (as investigated above), many customers seek to address this uncertainty or risk by including text in the contract to the effect that certain designated types of loss will be ‘deemed’ to be direct losses (so as to be certain to be claimable by the customer in the event that such loss arises). 

Some customers seek to justify the listing of such losses (which typically include internal management costs, additional procurement costs etc) by saying that the clause is a ‘for the avoidance of doubt’ type provision, as the losses in question would likely be classified as direct losses anyway.  However, such an argument is ultimately flawed, as of course if such losses were, in the circumstances, all ‘direct’ losses in the eyes of a court, then they would be properly claimable in any event.  

The better argument is simply that, if a dispute were to arise, a lot of legal time (and expense!) might be spent in seeking to argue the toss as to whether the loss in question should be treated as direct or indirect, with the defendant (normally the supplier) doing its utmost to muddy the waters in the hope that the amount of any eventual settlement of the claim might be as low as possible. While the recent decision in GB Gas Holdings Ltd v Accenture (UK) Ltd [2010] EWCA Civ 912 saw the Court of Appeal favour a narrower interpretation of indirect loss, the case did not remove the inherent uncertainty around the meaning of these words. This degree of uncertainty can undermine the business case of one and possibly both parties. Therefore, it is better to have the debate pre-signature. This allows the parties to identify where there is no consensus and, if necessary, facilitates an adjustment being made either to the contract or to the business cases which are affected. 

This approach may accordingly be justifiable, although much will depend on how far the scope of the ‘deemed direct’ losses gets pushed; the more the losses in question amount to the type of losses which would, almost certainly, have been said by the courts to be a ‘direct’ loss in any event, the more appropriate this approach is.  

However, what is less supportable is some of the wider formulations of this type of clause which we have seen in recent times, and which plainly include types of loss which would more properly be described as indirect, or even too remote to be recoverable at all (which would then serve to undermine some of the other limitation or exclusions in the contract). 

7.         Conclusion

It seems to us that, in the area of limitation of liability, relying too heavily on standard drafting and arguments risks (a) undermining the protection sought by each party and (b) producing a contract with results which may leave a nasty surprise when reliance is placed upon its terms. Time spent at the outset in ensuring that there really is a consensus around when and how the limitations of exclusions apply is likely to be amongst the most productive throughout the whole negotiation process, and will ensure that the contract really does balance the rights and interests of both parties fairly. 

Kit Burden and John McKinlay are Partners at DLA Piper(Technology, Sourcing & Commercial Group).