Introduction

Despite the saying that 'a verbal contract is not worth the paper it is written on' (attributed to the American film producer Samuel Goldwyn), there is no shortage of cases in the English courts based on alleged oral agreements, or oral variations of written contracts. Such claims may well face evidential difficulties, and they might ultimately fail. But they inevitably introduce risk and uncertainty for the party resisting such assertions, who will be exposed to legal costs. Leaving alleged oral understandings to one side, another way in which written agreements may be varied is by subsequent conduct. One regularly encounters situations where after signing the contract, the parties both fall into the habit of performing it in a manner that is inconsistent with its terms.

To protect the written bargain, most commercial agreements contain an anti-oral variation clause as part of the 'boilerplate' clauses. Anti-oral variation clauses state that no variation shall be effective unless it is in writing and signed by the parties through their duly authorised representatives. For a number of years, it was unclear whether such provisions were effective under English law, with apparently conflicting Court of Appeal decisions on the point. 

Now, in Globe Motors Inc v TRW Lucas Variety Electric Steering Ltd [2016] EWCA Civ 396, the Court of Appeal has provided the most detailed consideration of the authorities, both for and against, so far. Although obiter, the Court of Appeal concluded that anti-oral variation clauses were flawed as a matter of basic legal principle. That view may prove persuasive in future cases. This article considers the authorities leading up to, and the decision in, Globe Motors, and outlines how the parties nevertheless might protect themselves against claims based on an oral variation or implicit understanding. 

The case for: United Bank Ltd v Asif (Court of Appeal, 2000)

Parties relying on anti-oral variation clauses tend to cite the Court of Appeal decision in United Bank Ltd v Asif. In that case, in January 2000, the Court of Appeal denied permission to appeal from a summary judgment against a guarantor, who had been held liable to repay a loan of £6 million on which the debtor had defaulted. At first instance, the guarantor had advanced a defence based on an alleged oral variation of the guarantee. He claimed that the bank had (orally) agreed to extend time for payment, and (even better) had also significantly reduced the amount that was guaranteed. 

The deed of guarantee provided that "no variation ... shall be valid or effective unless made by one or more instruments in writing signed by the parties ...". When granting summary judgment for the bank, Wright J kept his reasoning brief. He rejected the submission that an officer of the bank had simply disregarded the anti-oral variation clause in the deed. The judge also found that the allegations of fraud (described as 'somewhat Byzantine' on appeal) were unsubstantiated, and that the bank's officer accused of letting the guarantor off the hook had lacked the authority to bind the bank. On appeal, Sedley LJ upheld the judge's decision "that no oral variation of the written terms could have any legal effect". He also found (separately) that the bank's officer had not had actual or ostensible authority to make the concessions in any event. Neither the judge nor the Court of Appeal however explained why the anti-oral variation clause was itself effective to preclude any variation: it seems the point had not been argued in any kind of detail. Despite the brevity of the Court of Appeal's judgment, United Bank Ltd v Asif has subsequently been treated as authority for the proposition that these clauses do work. 

Before moving on to the next Court of Appeal decision on anti-oral variation clauses (which went the other way), it is worth noting that the alleged oral variation in United Bank also failed because the relevant officer of the bank who had allegedly acceded to the variation, acting as agent of the bank, had lacked any authority to bind his principal in this manner. This point takes on greater significance for parties in the wake of Globe Motors v TRW, as will be seen.

The case against: World Online Telecom Ltd v I-Way Ltd [2002] EWCA Civ 413

Two years later, the Court of Appeal heard another appeal from a summary judgment decision, in World Online v I-Way. The dispute concerned the payment that I-Way, an internet service provider, would receive from World Online in return for providing hardware necessary to ensure that World Online's customers would be able to access the internet. The contract was signed in April 1999. I-Way argued that as early as May 1999, it became clear that the hardware requirements to enable internet access were in fact more extensive than had been anticipated, so that I-Way asked for, and was granted, an additional 10% of the relevant amount from World Online. I-Way said that an oral agreement had been concluded to that effect, and that the extra 10% was then paid regularly by World Online.

World Online applied for summary judgment in reliance on an anti-oral variation clause, which provided that:

"... no addition, amendment or modification of this Agreement shall be effective unless it is in writing and signed by and on behalf of both parties."

The judge refused summary judgment. He found that I-Way had a reasonable prospect of overcoming the provision, and neatly summed up the logical conundrum that casts doubt on whether these clauses bite. Irrespective of such a clause, in theory parties are free to conduct themselves as they wish, and enter into a new agreement:

"... orally or by conduct, for that matter, if they chose. That agreement can either be treated as an oral agreement varying the original agreement, or as a free-standing contract, and I know nothing in case law that prevents such an event having effect."

On appeal, Sedley LJ found that the judge was right to have ordered a full trial. Having had the benefit of argument on the point, his view was that "... the parties have made their own law by contracting, and can in principle unmake or remake it ...". Reaching a subsequent oral agreement notwithstanding the clause in the original contract would therefore be an exercise of freedom of contract - a fundamental principle of English law. Ultimately, the Court of Appeal noted that the law was not fully settled on the point, and that alone was a good enough reason for having a trial. 

The matter did indeed go to trial before David Steel J (I-Way Ltd v World Online Telecom Ltd [2004] EWHC 244 (Comm)). The judge upheld I-Way's claim that the contract had been varied orally in the course of a meeting, which (as he found) had ended with a consensus and a handshake. The judge did not refer to the anti-oral variation in the judgment, which suggests that World Online was either not pursuing that line of argument, or (at least) had not been pushing the point. It is worth noting that the oral agreement (which the judge upheld) had subsequently been referred to in correspondence and minutes of other meetings, although World Online never seems to have unequivocally accepted it in writing. So there was, at least, some 'written' evidence of the agreement, though not in the form of a duly signed amendment.

Upping the stakes

The next case in which anti-oral variation clauses featured upped the stakes financially. In Energy Venture Partners Ltd v Malabu Oil and Gas Ltd [2013] EWHC 2118, Gloster LJ (sitting as a judge of the Commercial Court) had to decide whether Energy Venture Partners ("EVP") and Malabu had orally varied their written contract such that Malabu had to pay EVP an amount of up to US$ 200 million, in return for services provided by EVP in relation to the sale of Malabu's interest in an oil field in Nigeria. Essentially, under the written agreement, EVP was to act exclusively for Malabu as a broker. EVP would seek to obtain offers for the block in question. EVP would then be paid a commission by Malabu if there was a qualifying sale following an offer procured by EVP during the exclusivity period. 

This was a complex dispute, and Gloster LJ's judgment runs to 337 paragraphs. She found that Malabu had ultimately entered into a transaction that, in principle, entitled EVP to a fee under the contract. However, it was not at all clear what that fee should be. That was because the parties had, apparently quite deliberately, departed from the contractual provisions setting out how EVP's commission was to be calculated. The contractual mechanism required an agreement between Malabu and EVP as to a particular price, which then governed how much EVP would receive. Malabu did not agree that price, so it became impossible to calculate EVP's commission based on the written contract. 

Gloster LJ found that the parties had reached an implied agreement, orally at a particular meeting and by conduct thereafter, that EVP should nevertheless be paid a fee for its services. She went on to imply a term into the agreement that EVP was entitled to a 'reasonable' fee (which turned out to be around US$ 110 million). In resisting the claim based on an implied agreement, Malabu had relied on an entire agreement clause, which said that "... no additions, amendment to or modifications of this Agreement shall be effective unless it is in writing duly signed on behalf of the parties ...". Malabu accepted that this clause could subsequently be overridden if the parties in fact entered into an oral agreement (and EVP agreed with that proposition), but nonetheless submitted that "... the purpose of such clauses, is not to prevent the recognition of oral variations, but, rather, [to prevent] casual and unfounded allegation of such variations being made." 

Gloster LJ noted that neither party had relied on any authorities holding that such clauses were ineffective as a matter of law, and that the enquiry would therefore have to be fact based. She concluded that while this was not the occasion to decide the point of legal principle:

"... as at present advised, I incline to the view that there can be an oral variation in such circumstances, notwithstanding a clause requiring written modifications, where the evidence on the balance of probabilities establishes such variation was indeed concluded."

That conclusion was tantamount to saying that anti-oral variation clauses were of no effect, since the burden of proof for establishing an oral agreement would also be 'on the balance of probabilities' absent any such provision. On the facts before her, Gloster LJ found that an implied agreement had been proven, regardless of the anti-oral variation clause.

No more appetite for anti-oral variation clauses?

By then, parties appeared to have lost their appetite for relying on anti-oral variation clauses being effective to preclude unwritten variations or agreement. In Virulite LLC v Virulite Distribution Ltd [2014] EWHC 366 (QB), the High Court again proceeded on the agreed assumption that such a clause could not preclude any subsequent waiver, estoppel or indeed variation from arising as a matter of law. Instead, the argument was that anti-oral variation clauses gave rise to an 'evidential presumption' that the parties did not intend to vary the contract unless they produced a signed, written instrument to that effect. In effect, it was said that these clauses made it more difficult to establish the oral variation.

Stuart-Smith J rejected that argument. He noted that the concept of an 'evidential presumption' was not something that had been sanctioned by authority under English law, and that the standard of proof remained the balance of probabilities. References to the need for 'strong evidence' were unhelpful, as the Court would (as it ordinarily does) consider all the evidence, and giving due weight to all the facts.

The judge did note, however, that the parties having included such a clause in their agreement was a relevant factor, and the circumstances in which such a clause was included in the contract in the first place might also be relevant. The existence of an anti-oral variation clause would require the Court to "look closely" at whether a subsequent agreement was actually concluded and "whether, in reaching that agreement, the parties had intended to enter into legal relations so as to vary the terms of their original contractual obligations."

The judge's second point, relating to the intention to enter into legal relations, merits further commentary. Together with offer and acceptance, and consideration, an intention to create legal a relationship is one of the fundamental requirements for a binding contract under English law. In the commercial context, where contracts (as opposed to letters of intent or documents marked 'subject to contract') are being negotiated, an intention to create legal relations is usually found. Indeed, this requirement often goes unmentioned as it has so obviously been met. 

As will all matters relating to contract formation, English law adopts an objective approach to determine whether there was the necessary intention. In RTS Flexible Systems Ltd v Molkerei Alois Muller [2010] UKSC 14, the Supreme Court summarised the general principles:

"... Whether there is a binding contract between the parties and, if so, upon what terms depends upon what they have agreed. It depends not upon their subjective state of mind, but upon a consideration of what was communicated between them by words or conduct, and whether that leads objectively to a conclusion that they intended to create legal relations and had agreed upon all the terms which they regarded or the law requires as essential for the formation of legally binding relations. Even if certain terms of economic or other significance to the parties have not been finalised, an objective appraisal of their words and conduct may lead to the conclusion that they did not intend agreement of such terms to be a pre-condition to a concluded and legally binding agreement."

As the Court of Appeal noted in Barbudev v Eurocom Cable Management [2012] EWCA Civ 548:

"In a commercial context, the onus of demonstrating that there was a lack of intention to create legal relations lies on the party asserting it and it is a heavy one."

An anti-oral variation clause in an existing contract between the parties is certainly an 'objective' matter. A party resisting a claim that the contract was varied notwithstanding such a provision should therefore be able to rely on it in denying that it lacked the requisite intention to alter the agreement. That is what Stuart-Smith J said in Virulite v Virulite. However, the learned judge also held that the mere existence of the clause did not somehow shift the onus on to, or increase the burden of, the party seeking to overcome the clause. 

The end of the road? Globe Motors v TRW

The Court of Appeal in Globe Motors Inc v TRW Lucas Variety Electric Steering Ltd [2016] appears to have now put the final nail in the coffin of anti-oral variations clauses. The appeal concerned a dispute under a long term agreement for the exclusive supply of electric motors and certain other components used in power assisted steering systems for cars. TRW, the defendant, assembled these power steering systems. Under the supply agreement, TRW agreed to purchase all electric motors and lead frame assemblies that it might require exclusively from Globe Motors. The exclusivity obligation in the contract only bit, however, if the relevant motors and assemblies fell within the specific scope of what was called the 'Products' under the agreement. 

The parties entered into their supply agreement in 2001. In or around 2005, TRW began to purchase certain 'second generation' motors from a company called Emerson. In 2006, TRW then acquired Emerson, and Emerson continued to supply the next generation motors to TRW. By 2014, TRW had purchased more than 3 million units from Emerson. Globe Motors alleged that this was a breach of the exclusivity obligation in the supply agreement. TRW argued in its defence that the second generation motors were not in fact 'Products' for the purposes of the supply agreement. TRW said this was because the contract provided for a particular process by which engineering specifications could be altered so as to widen or extend the scope of the 'Products', as originally defined (so-called Engineering Changes): however, this contractual machinery had never been operated with regard to the type of motor that TRW had been buying from Emerson. 

At first instance, the judge decided that the exclusivity obligation applied to the second generation motors, because they were a product that Globe 'could and would have built' if TRW had required that of Globe pursuant to the agreement by operating the machinery - something that TRW could have done. The judge relied on the fact that the Engineering Changes procedure could have applied to the Emerson motors, and if that had happened, Globe would have had to buy all of its requirements from TRW under the exclusivity arrangement. He concluded that any new type of motor that "could and would" have gone through the Engineering Changes process was also a Product (or, at least, was deemed to be one). Globe succeeded at trial.

The Court of Appeal held that the judge had fallen into error. The key reason for allowing TRW's appeal was that under the contract, Engineering Changes had to be initiated by TRW. Globe would then have to agree to these changes, always provided it was capable of manufacturing the new specification. However, there was no obligation (express or implied) on TRW to go through the Engineering Changes process with Globe, once TRW had realised that it might need a different type of motor to which the contract did not (yet) apply. 

In finding for the claimant, the judge had been influenced by the fact that the supply agreement was a long-term, 'relational' contract that was clearly of great importance for Globe. Globe had invested and increased its manufacturing capabilities specifically to allow it to perform the contract. The Court of Appeal noted that in some long-term contracts, there might be scope for the implication of a duty of loyalty or good faith, or a duty to co-operate: such a duty might be breached by one party deliberately refusing to operate contractual machinery that was plainly intended to apply, so that a new product would not be covered by the contract. Beatson J was however careful to point out that any implication of such a duty of good faith would only succeed where the contractual wording permitted it, and would not follow simply because the contract fell into a particular category (such as a long-term, exclusive supply agreement):

"... an implication of a duty of good faith will only be possible where the language of the contract, viewed against its context, permits it. It is thus not a reflection of a special rule of interpretation for this category of contract." 

In this case, Globe might, nonetheless, have succeeded with an argument that there was an implied duty:

"In the present case, where the Agreement was for exclusive supply and, absent insolvency or material breach, was for the lifetime of the platforms and a long term agreement, the flexibility of approach I have described might have given considerable force to a submission that there was an implied obligation on TRW Lucas to give Globe an opportunity to show that it could provide a Gen 2 motor."

Unfortunately for Globe, any claim for a breach of such an implied obligation would have been time-barred, as more than six years had elapsed since TRW awarded the contract to Emerson. Globe therefore failed to achieve by a process of interpretation of the express terms of the contract what it might have been able to achieve by a process of implication. The express terms simply said that it was up to TRW to initiate the Engineering Changes.

Having found that the judge was wrong to conclude that TRW was in breach of contract, the Court of Appeal did not need to decide another ground of appeal, which concerned the effectiveness of an anti-oral variation provision contained in the supply agreement. The judge had also found that an affiliate of Globe's had become a party to the contract (in addition to the named parent company), by conduct or through an implicit understanding between the parties, notwithstanding the anti-oral variation clause. Based on that finding, the judge awarded damages for losses suffered by Globe's affiliate - which could not stand as the appeal was allowed. 

Nevertheless, Beatson, Underhill and Moore-Bick LJJ all offered obiter conclusions on whether the anti-oral variation clause could be effective as a matter of principle, having had the benefit of full argument. As will be seen, they decided against such clauses being effective. Before turning to the reasons for that, it is worth looking at just how the parties had (according to the judge) varied their contract and added a party.

Globe had a subsidiary, incorporated in Portugal (referred to in the judgments as "Porto"). Porto had been established with the intention of supplying motors to TRW under the supply agreement. While Porto was not a named party to the supply agreement, Globe argued that the contract was subsequently varied so as to make Porto a party. That was because Porto had in fact performed the agreement on behalf of Globe, and had been supplying motors to TRW from November 2002 onwards. Globe relied on the following matters as having led to Porto becoming a party to the contract:

- From early 2003, TRW had placed orders with Porto for 'Products', in accordance with the contractual specifications and prices. Porto had in turn supplied the 'Products', in line with the contract. Porto also invoiced TRW (or its nominees), and payments were made to Porto.

- TRW submitted warranty claims made under the applicable provisions of the supply agreement to Porto. TRW also submitted volume forecasts (as envisaged by the contract) to Porto. None of the relevant contractual correspondence or notices were addressed to Globe.

- In 2005, TRW and Porto (not Globe) entered into a supplemental agreement, in order to satisfy certain consignment stock requirements set out in the supply agreement.

On these facts, the judge found that it was overwhelmingly clear that the parties had intended to vary the supply agreement, such that Porto had become a party to it. Pausing here, one can see that a contrary conclusion might seem unfair and artificial. If Globe had really wanted to keep the contract at parent company level, then the parent entity could and should have taken the lead in managing and performing the agreement.

That left the question whether the anti-oral variations clause precluded that subsequent variation from being effective and binding as matter of law. Article 6.3 of the contract provided:

"6.3 Entire Agreement; Amendment: This Agreement, which includes the Appendices hereto, is the only agreement between the Parties relating to the subject matter hereof. It can only be amended by a written document which (i) specifically refers to the provision of this Agreement to be amended and (ii) is signed by both Parties."

Beatson LJ rejected the submission that the Court of Appeal was bound by the prior decision in United Bank v Asif. He took the view that, on its face United Bank v Asif (though it was a brief judgment) was inconsistent with World Online Telecom Ltd v I-Way Ltd. As there were two inconsistent decisions at the appellate level, the incarnation of the Court of Appeal that had to decide Globe Motors v TRW was free to choose which prior judgment it would follow. As the Court of Appeal in United Bank had only approved the judge's proposition that these clauses were effective, without giving any reasons, Beatson LJ preferred to follow the later judgment in World Online, which was also in line with the other first instance decisions set out above (Energy Venture Partners and Virulite).

The Court of Appeal noted that anti-oral variation clauses were meant to serve a purpose, of preventing parties having to incur costs in defending ill-founded allegations of oral agreements, whilst at the same time promoting certainty in commercial agreements. However, these considerations could do nothing to alter the position as a matter of legal principle, that:

"Absent statutory or common law restrictions, the general principle of the English law of contract is that ... The parties have freedom to agree whatever terms they choose to undertake, and can do so in a document, by word of mouth, or by conduct. The consequence in this context is that in principle the fact that the parties' contract contains a clause such as Article 6.3 does not prevent them from later making a new contract varying the contract by an oral agreement or by conduct."

Beatson LJ also agreed with Gloster LJ's comments in Energy Venture Partners, that it was undesirable to attach any adjectives or qualifications to the evidence that a party seeking to prove an oral variation in the face of such a clause would have to adduce: the question was simply whether on the balance of probabilities, they had established a subsequent agreement.

Underhill LJ added that it did not, however, follow that anti-oral variation clauses were of no value at all:

"In many cases parties intending to rely on informal communications and/or a course of conduct to modify their obligations under a formally agreed contract will encounter difficulties in showing that both parties intended that what was said or done should alter their legal relations; and there may also be problems about authority. Those difficulties may be significantly greater if they have agreed to a provision requiring formal variation."

Conclusion

The principle of freedom of contract provides an absolute answer to any suggestion that anti-oral variation clauses can preclude a subsequent oral agreement from being effective in law. If the parties can agree today that they must amend their contract in writing, they can also agree tomorrow that they no longer want to be bound by this restriction, and that they are prepared to amend their contract orally. If the parties entered into an oral variation, and thereafter performed the contract as if it had been duly amended, there may also be a case for an estoppel preventing either of them from denying the oral variation (assuming the necessary reliance and detriment were present on the facts). 

While black-letter lawyers everywhere will rejoice, the position is less satisfactory from a practical point of view. Underhill LJ in Globe Motors v TRW clearly recognised that allegations of oral variations create uncertainty and make the contract more difficult to enforce, because there may be enough in such claims at first blush to have the matter go to a full hearing. Having looked for ways in which anti-oral variation clauses might be given effect, he could find none that were doctrinally sound - and neither could Moore-Bick LJ, the third member of the Court of Appeal.

Nonetheless, it is suggested that there are ways in which these clauses can be tightened up such that they might still carry the day on a summary judgment application in Court (or be enough to have an arbitral tribunal reject an 'oral variation' as a preliminary matter). The key lies in the fact that parties to commercial contracts are almost always companies, who act through duly authorised agents. If the scope and limits of an agent's authority are clearly set out and explained to the counterparty, which can be drafted into the anti-oral variation clause, then it becomes more difficult to argue that the principal is nevertheless bound where the agent has exceeded their actual authority (by operation of the law of agency as to 'ostensible' or 'implied' authority). 

Certain contracts in the construction and energy sector appoint named representatives and make them solely responsible for the administration of the contract. There is no reason why an anti-oral variation clause could not, in addition, refer to a named individual, ideally a director of the company, as having the sole authority to amend or alter the written agreement, and go on to state that any successors must be duly appointed by the board and be notified in writing to the other party. The counterparty (or at least its corporate ego) would then be expressly aware of who is authorised to bind the company in question - and an analogy can be drawn with the corporate counterparty having the requisite intention to be legally bound. Any oral agreement that could vary a contract that includes this additional protection then becomes more involved, and hence more difficult to prove: not only must there have been a new agreement on the substance of the variation, but it must also have been made by the relevant director - and preferably in writing (though as we have seen, lack of writing alone is not decisive). At the very least, amending these clauses in this way gives a company resisting an oral variation another string to its bow, based on a lack of authority.

The content of this article is intended to provide a general guide to the subject matter. Specialist advice should be sought about your specific circumstances.