Pinsent Masons provides legal advice in breakthrough data trust report for food industry
Head of Data Law Solutions, Netherlands
Out-Law Guide | 02 Nov 2007 | 8:31 am | 12 min. read
Phoenix Media Limited v Cobweb Information Limited
The parties had entered into an agreement dated 29 November 1999 by which Cobweb licensed Phoenix to incorporate its products (basically, a compilation of business information from a wide variety of sources) as the core element of a wider business information package marketed by Phoenix as “IQ, the ultimate business information service”. This would be marketed to “subscribers”.
This agreement replaced an earlier agreement, and tried to deal with what were perceived as shortcomings in the previous commercial arrangements. Firstly, the new agreement re-defined “subscribers” as entities as having less than 400 employees (rather than the old 50); secondly, it allowed Phoenix to sub-license the Cobweb product to distributors who could themselves sub-license the product to subscribers and finally, Phoenix paid Cobweb not only a percentage of subscription fees but also of sponsorship, advertising and other revenue received by Phoenix and its distributors.
Perhaps it was at least partly because the new agreement altered the existing contract that the new contract was of frightening complexity. Indeed, the judge commented at several instances on the byzantine drafting, involving a consideration of numerous interlinked clauses and schedules to the agreement. Much of the judgment consists of a lengthy exposition of the most important provisions, which it is not intended to recite here. Suffice it to say that the scheme of the new contract was to allow Phoenix to sell the combined product incorporating Cobweb's product only to “subscribers” with fewer than 400 employees - small and medium sized enterprises.
The important termination provision, however, needs to be read in its entirety:
"Either party may terminate this Agreement forthwith by notice in writing if the other of them is in material breach of this Agreement and shall have failed (where the breach is capable of remedy) to remedy the breach within 30 days of the receipt of a request in writing from the party not in breach, to remedy the breach such request setting out the breach and indicating the failure to remedy the breach may result in termination of this Agreement.”
Cobweb complained in early February 2000 that Phoenix had breached the agreement by approaching organisations that were in excess of the 400 employee limit. This was followed by a letter on 14 February purporting to terminate the agreement. It was in these terms:
"We also understand that [Phoenix] has dealt directly with Daily Mail, Spicers, Norweb, Hewlett Packard, MCI/Worldcom and Lloyds TSB. None of these companies is capable of being either a Distributor or Subscriber under the Agreement. Each of these companies is an Excluded Organisation as defined in the Agreement. Your dealings with these companies amount to breaches of clauses 2.4, 2.5, 7.4, 7.6 and 15 of the Agreement. In light of the above we consider [Phoenix] to be in material breach of the Agreement on a number of counts. Furthermore, the majority, if not all, of the breaches outlined above are incapable of remedy.”
The matter came before the judge after an adjournment to permit a mediation, which had failed to bring about a compromise. Phoenix applied for declaratory and injunctive relief with regard to the purported termination by Cobweb and applied for summary judgment. The parties agreed that the matter should be finally disposed of by the judge, and the evidence consisted of witness statements by a representative of each party with exhibits.
The question therefore remained, given that there were breaches, were they material and irremediable? The clause on termination required that Cobweb could only have terminated the agreement if the breaches were both material and irremediable. So there could be no right to seek to terminate if the breaches were not material, and the right to terminate forthwith could not exist unless the breaches were irremediable. If they were remediable, 30 days' notice had to be given, and if the breaches were remedied within that period, that would be the end of the matter.
For materiality, Neuberger J. thought that the court should have regard to the actual breaches, the consequences of the breaches to Cobweb, Phoenix' explanation for the breaches, the breaches in the context of the agreement, the consequences of holding the agreement determined, and the consequences of holding that the agreement continued.
Neuberger J. found that the actual breaches were that there had been negotiations between Phoenix and some “Excluded Organisations” for them to act as sponsors, possibly as distributors. The product had been provided in some cases to some of these organisations for the purpose of evaluation.
As to the consequences of those breaches, the judge observed that the contact between Phoenix and the “Excluded Organisations” had not gone beyond preliminary discussions. Where the product had been supplied, there would have been at least implied obligations of confidentiality, and if the negotiations had come to nothing, the product would have been returned. While there might have been some risk of damage, this could have been compensated by damages. Phoenix could have called off the negotiations and asked for the products to be returned. Accordingly, this would not be a material breach.
There was a dispute as to whether Phoenix acted dishonestly. The judge found no such dishonesty on the basis of the evidence he had seen. “Pretty clear evidence” would be required before such a finding, and the judge took account of the fact that there had been no cross-examination of Phoenix' witness.
The judge compared the consequences of the agreement being terminated or continuing. If terminated, the consequences for Phoenix would be severe, with a large part of the considerable amount of money invested by it in the product being wasted (or Phoenix would be at Cobweb's mercy in any re-negotiations). There were no such bad implications for Cobweb if the agreement continued, even taking account of the bad feeling that had arisen between the parties.
A final point to take into account was that the breaches took place in an agreement that was to last for 3 years and possibly over 8 years.
The question of remediability was related to materiality, and there was some overlap. However, the judge was of the view that the breaches were remediable. Unlawful negotiations could be called off, and the products had been supplied under at least implied obligations of confidentiality and they could have been returned within the 30 days.
There was a dispute as to whether Phoenix would have put the breaches right: it may have been that Phoenix would have persisted in its alternative interpretation of the agreement and done nothing. The judge regarded this as “irrelevant speculation”. The breaches were capable of remedy and so Cobweb first had to require their remedy within 30 days, and then had to establish that they remedying had not been done in order to justify termination. Cobweb did not require the remedying and as they had not established that the breaches were irremmediable, the agreement was not terminated, even if the breaches were material.
Termination clauses are often treated by draftsmen as an extension of boilerplate clauses, and all too often they are simply taken from a precedent contract and then inserted into a draft contract with scarcely a thought as to what they might mean in practice. The type of termination provision which this case was concerned with is one which might grace almost any commercial contract in this jurisdiction, but how many commercial lawyers ask themselves the simple question, “in the context of the sorts of breach that might happen in this particular type of deal, what sorts should give rise to a right to termination and what sorts of breach are, or are not, remediable?”
Termination is always a difficult question in any agreement that is likely to change over time - and this includes in the IT context, outsourcing, distribution, and major systems development or integration projects. It is with these sorts of contracts that changes are most likely to take place, both in the context of formal change control, and also “informal” change control, in the myriad of little agreements that the parties make in the ordinary course of working the contract. Sometimes, it is these little agreements that can have a major impact on the future operation of the contract, but their status outwith formal change control can make them hard to identify.
This is typically the case with major systems developments. There will be formal, signed off change control, and also other variations to the agreement that perhaps never went through the contractual mechanism, which may have resulted from agreements reached at requirements analysis workshops or at Project Board level. Similar points can be made about outsourcing or facilities management contracts. It is the fact that they were not part of formal change control that makes them so hard to identify, but many of them undoubtedly do vary the formal terms of the contract, on important issues, such as timescales. When called on to advise with regard to termination for delay, for example, all these agreements have to be taken into account in assessing whether the delay complained of amounts to a “material” breach or not, seen against the varied deadlines.
However, the story does not end there. Assuming that a proper job has been done of re-constructing the contract in the light of all the formal and informal change control, what is the proper test to apply to determine if the victim of breach of contract can elect to terminate? Applying the test of “repudiatory breach” is immensely difficult in these cases. It is now quite some time since the courts simply countenanced the existence only of “conditions” (whose breach would give the other party the right to terminate) and “warranties” (whose breach would give the other party the right to damages only). This has all changed with the introduction of the concept of “innominate” terms and this principle has been developed since Hong Kong Fir Shipping Co Ltd v Kawasaki Kisen Kaisha Ltd  2 QB 26. Lord Wilberforce in Bunge Corporation v Tradax Export SA  1 WLR 711 said that Diplock LJ in Hong Kong Fir:
"...illuminated the existence in contract of terms which are neither, necessarily, conditions or warranties, but in terminology which has since been applied to them, intermediate or innominate terms capable of operating, according to the gravity of the breach, as either conditions or warranties."
This sort of consideration was recently considered in the context of what was effectively a facilities management contract in the case of Rice v Great Yarmouth Borough Council. Rice had a four year contract for the maintenance of Great Yarmouth's grounds and sports facilities. Rice won against a bid from Great Yarmouth's own services organisation. Part of Rice's case was that many of his difficulties could be attributed to the reluctance of the council's workforce to transfer to his employment and also the reservations held by the council about using Rice, as opposed to the council's own services organisation.
Notices of breach from the council followed quickly, some were proved, some were not, others were trivial. The council purported to terminate relying on a provision in the agreement allowing termination if the contractor committed “any breach of its obligations” under the contract. Lady Justice Hale gave the judgment of the Court of Appeal and upheld the trial judge's finding that a sensible, commercial construction had to be given to the termination provision. It could not be the case that any trivial breach could be a justification for termination but it had to be interpreted to mean that only a repudiatory breach or an accumulation of breaches that as a whole could be described as repudiatory could be within the termination provision. The parties could always specifically provide that a term was so important that any breach would justify termination, but that was not the case here. In other words, in such contracts, the court could read in “material” into the contract's termination provisions even though it was absent from the text.
In Rice, the judge did not think the contractor's behaviour could be categorised as “repudiatory”. The Court of Appeal agreed. It was not in all cases necessary to demonstrate an intention to abandon contractual obligations and Hale LJ said that there are really in this sort of case three categories of “repudiatory breach”:
"(1) those cases in which the parties have agreed either that the term is so important that any breach will justify termination or that the particular breach is so important that it will justify termination; (2) those contractors who simply walk away from their obligation thus clearly indicating an intention no longer to be bound; and (3) those cases in which the cumulative effect of the breaches which have taken place is sufficiently serious to justify the innocent party bringing the contract to a premature end."
The important feature of such contracts is that they run over a course of time, and it is possible to look at breaches cumulatively and see what they say about the contractor's ability to perform in the future. It is also permissible to look at the surrounding circumstances – such as evidence of the council's behaviour and the effect of the drought on Rice's performance. The question to ask was whether the council had, in all the circumstances, been deprived of a substantial part of the totality of that which it had contracted for that year, subject to the additional possibility that some aspects of the contract were so important that the parties are to be taken to have intended that depriving the council of that part of the contract would be sufficient in itself.
These words of the Court of Appeal, albeit in the context of a local authority contract for facilities management, are of immediate relevance to the IT industry and are important for an understanding of Neuberger J.'s reasoning in Phoenix. It is not permissible to look just at the contract's literal terms (after taking into account change control) but also to look at the whole commercial environment of the contract and of the breaches. These factors are well described in both Phoenix and Rice.
Taking the example of an outsourcing contract which is performed well below the levels in the Service Level Agreement, it is not in every case that this would justify termination even for such apparently serious failings. A court would look at the reasons for the breaches, what they said about future conduct of the agreement, the effect on the victim of the breach and so on.
A word should be said about “remediable”. It occurs so frequently in commercial contracts that few give any thought to what it means. What types of breaches are remediable and which are irremediable?
Neuberger J. gave the common example of a clause in a lease for doing decoration in the fifth year. In a sense, this is not remediable after the end of the fifth year: without a time machine, it is not possible to go back in time and do the decoration in the fifth year. It is, of course, possible to do the decoration in the sixth or subsequent years and so, in common sense, the breach is remediable. As the judge observed, although the negotiations had started with “Excluded Organisations”, it was possible to stop them, and products delivered could be recalled. In this sense, such breaches were “remediable”.
The judge did not cite, and it is not apparent that the authority was provided to him, the case of Glolite Limited v Jasper Conran Limited the Times 28 January 1998 on this subject. This is all the more surprising since it is also a decision of Neuberger J. Glolite also looked at “remediable” in the context of a long-term agreement. There was a ten year agreement (extendable to twenty years) for the manufacture of certain items designed by the defendant for sale to retail and wholesale outlets in most of the world. The parties were to agree on the publicity aspects from time to time, and the defendant claimed that this had not happened. The termination provision read:
“[either party] may terminate this agreement forthwith by notice to the other if the other party commits any material or irremediable breach of any term of this agreement or (if such breach is capable of remedy) has failed to remedy it within 28 days of a notice giving adequate particulars of the alleged default.”
Where breaches had taken place, it was clear that the past could not be undone, but on the facts the unauthorised publicity had not been extensive and there was no evidence of harm to the Jasper Conran reputation. The judge accordingly held that there was no “material and irremediable” breach. Furthermore, the fact that the clauses breached were very important to one party did not mean that any breach would be material. The “remedy” was in not making any more shirts with the offending logo on them and stopping the publicity.
Glolite did cite other authorities, and those interested should certainly look at Schuler v Wickman Tools  AC 235 where the House of Lords also thought that “remedy” meant cure for the future rather than putting right past breaches.
Phoenix is a perfect example of two things. Firstly, the fact that advising on termination in the context of contracts that are supposed to go on over a long period of time is so very difficult and needs to take account of so many factors. Secondly, maybe it is time for boilerplate termination provisions to be looked at anew and some searching questions asked (and answered) about whether they are suitable for the contracts in which they are so frequently inserted without further thought.
Pinsent Masons provides legal advice in breakthrough data trust report for food industry
Head of Data Law Solutions, Netherlands