Incremental Steps Towards Clarity: Good Faith as an Organising Principle Behind the Implication of Terms in Fact

I. Introduction

The general reluctance of the law of England and Wales to recognise a duty of good faith in contract performance is well documented. Rather, English law embodies an “ethos of individualism”,^[1] under which the parties are expected to act in rational pursuit of their own self-interest.^[2] This is perhaps most eloquently explained by Bingham LJ’s comments, in Interfoto Picture Library v Stiletto Visual Programmes:^[3]

In many civil law systems, and perhaps in most legal systems outside the common law world, the law of obligations recognises and enforces an overriding principle that in making and carrying out contracts parties should act in good faith. … English law has, characteristically, committed itself to no such overriding principle but has developed piecemeal solutions in response to demonstrated problems of unfairness.^[4]

One particular mechanism for the development of such ad hoc rules is the implication of terms in fact, based on the presumed intention of the parties – which, in Steyn LJ’s words, operate as “individualised gap-fillers, depending on the terms and circumstances of a particular contract.”^[5] One might, therefore, be forgiven for expressing some incredulity towards Leggatt J, in Yam Seng v International Trade Corporation,^[6] asserting there to be “no difficulty, following the established methodology for the implication of terms in fact, in implying [a duty of good faith into] any commercial contract based on the presumed intention of the parties”.^[7] Such a view seems contradictory not only to the general reticence towards the principle of good faith highlighted above, but even to the role of terms implied in fact themselves – which, as “individualised gap fillers”,^[8] cannot, by definition, have general application to “any commercial contract”.

A greater hurdle to the widespread acceptance of this notion, however, stems from subsequent developments in the field of terms implied in fact. As shall be discussed, Leggatt J’s approach draws heavily upon Lord Hoffman’s view, in Attorney General of Belize v Belize Telecom,^[9] that “there is only one question [in considering whether a term should be implied]: is that what the instrument, read as a whole against the relevant background, would reasonably be understood to mean?”^[10] More recently, however, the Supreme Court – in Marks & Spencer v BNP Paribas^[11] – has reaffirmed that the “traditional” tests for the implication of terms in fact must be satisfied, considering Belize not to substantively change the law of implied terms, but merely to provide authoritative guidance as to its doctrinal basis.

Leggatt LJ (as he is now) has more recently asserted – with little discussion – that the same outcome as in Yam Seng could be reached via the application of the traditional tests.^[12] This article will examine the validity of this assertion, identifying certain shortcomings in the application of the test. Despite this, it will be seen that there is a pervasive logic to the recognition of good faith as an “organising principle”, not substantively modifying but underpinning and informing the implication of terms in fact more generally; such recognition would constitute a relatively modest “incremental step”, in full accordance with Bingham LJ’s view of the “piecemeal approach” of the law of England and Wales. Finally, comparisons will be drawn with the approach of the Supreme Court of Canada in the landmark case of Bhasin v Hrynew^[13], in which such an incremental advancement was expressly undertaken.^[14] Further than this, however, the Court in Bhasin found that, “as a further manifestation of this organising principle of good faith, … there is a common law duty which applies to all contracts to act honestly in the performance of contractual obligations.”^[15] This finding goes somewhat further than simply underpinning and informing existing common law rules; accordingly, its compatibility with the law of England and Wales will be examined.

II. Yam Seng v ITC

Yam Seng v ITC itself related to a contract under which Yam Seng was granted the exclusive right to distribute certain products within the duty free markets of various regions around the world, as well as the domestic market of certain regions of China. In addition to a number of precontractual misrepresentations and breaches of the express terms of the contract, ITC engaged in a consistent pattern of dishonest communication throughout performance of the contract. This included: falsely claiming that an application had been made to register the products for sale in China; falsely claiming to have requested that the distributor in the domestic market of Singapore raise its prices so as not to undercut the prices in the corresponding duty free market; after eventually making that request, failing to disclose that the domestic distributor had immediately refused to do so; and, when the domestic distributor eventually agreed to do so, failing to disclose that this would take two months to implement. Yam Seng subsequently learned not only that the domestic prices had not been increased, but that the lower domestic prices were due to ITC’s own instructed pricing strategy. This led to the rapid deterioration of the parties’ working relationship, and, shortly thereafter, Yam Seng terminated the contract.

A key question for the Court was whether the consistent pattern of dishonest conduct breached an implied duty to act in good faith in the performance of the contract. In considering this submission, Leggatt J firstly noted there to be no general duty of good faith contract performance in the law of England and Wales – in particular, he drew attention to the “piecemeal approach” advocated by Bingham LJ, and three justifications for the “traditional English hostility” towards a principle of good faith highlighted by McKendrick:

The first is that English law starts from a premise of rugged individualism, in which the parties are expected to look after their own interests and to bargain and obtain the best terms which they can for themselves. … The second reason … is that English law is reluctant to embrace broad general principles, such as a duty of good faith. It prefers to develop incrementally and by analogy to existing precedents rather than by reference to broad statements of general principle. … The third reason, closely related to the second, is that a broad principle would generate too much uncertainty.^[16]

Thus, he was unhesitant in concluding there to be no general duty of good faith in contract performance.^[17] This does not, however, preclude the implication of a term in fact, based on the individual characteristics of the arrangement at hand; accordingly, he then outlined Lord Hoffman’s recharacterization of the implication process (discussed above) “as an exercise in the construction of the contract as a whole.”^[18] “As a matter of construction”, he noted, “it is hard to envisage any contract which would not reasonably be understood as requiring honesty in its performance.”^[19] This echoes the sentiment expressed both by Lord Hoffman himself – and even by Lord Bingham – in the context of construction more generally, in HIH Casualty v Chase Manhattan Bank.^[20] Based on Lord Hoffman’s approach to the implication of terms, he therefore appears justified in finding there to be “no difficulty, following the established methodology for the implication of terms in fact, in implying such a duty in any commercial contract based on the presumed intention of the parties”.^[21]

Despite the strength of this argument, finding there to be no difficulty implying the duty into “any contract” presents a contradiction – both with his aforementioned conclusion that there is no general duty of good faith, and (as noted at the outset) with the nature of terms implied in fact themselves, which operate as “individualised gap-fillers, depending on the terms and circumstances of a particular contract.”^[22] He did not, however, seem to strictly mean that such a duty could be found in “any” commercial contract – but, simply those that “require a high degree of communication, cooperation and predictable performance” (that is, the so-called “relational” contracts)^[23]. It would be erroneous, however, to consider Leggatt J’s “relational” contracts to constitute a discrete “category” of contract, into which a term of good faith performance might be implied: if this were the case, a term implied in law (which arises “as a necessary incident of a definable category of contractual relationship”)^[24] would be the more appropriate mechanism.^[25] Rather, the definition of “relational” contracts provided by Leggatt J should simply be regarded as listing the kinds of characteristics that might give rise to an enhanced duty of cooperation, based on the commercial expectation of honesty. This view is justified by noting that the “norms of behaviour” against which contracts are formed are, of course, highly sensitive to context; thus, the content of the duty itself must be equally contextual:

What good faith requires is sensitive to context. That includes the core value of honesty. In any situation it is dishonest to deceive another person by making a statement of fact intending that other person to rely on it while knowing the statement to be untrue. Frequently, however, the requirements of honesty go further. For example, if A gives information to B knowing that B is likely to rely on the information and A believes the information to be true at the time it is given but afterwards discovers that the information was, or has since become, false, it may be dishonest for A to keep silent and not to disclose the true position to B. Another example of conduct falling short of a lie which may, depending on the context, be dishonest is deliberately avoiding giving an answer, or giving an answer which is evasive, in response to a request for information.^[26]

In addition to this “core requirement”, Leggatt J noted a second, overlapping aspect of good faith: “what may be described as fidelity to the parties’ bargain. The central idea here is that contracts can never be complete in the sense of expressly providing for every event that may happen. To apply a contract to circumstances not specifically provided for, the language must accordingly be given a reasonable construction which promotes the values and purposes expressed or implicit in the contract.”^[27] This sentiment, of course, echoes Lord Hoffman’s words, in the context of terms implied in fact more generally, that “the need for an implied term not infrequently arises when the draftsman of a complicated document has omitted to make express provision for some event because he has not fully thought through the contingencies that may arise.”^[28] Thus, it appears that Leggatt J is essentially arguing that many, if not all, terms implied in fact – as well as the principles of construction more generally^[29] – fall within the broader umbrella of an implied duty of good faith. The validity and desirability of implying a broader, “umbrella” duty, encompassing a number of “lesser” implied terms, must therefore be examined.

III. Problems of Application

As noted, in reaching this decision, Leggatt J drew heavily upon Lord Hoffman’s approach to implication, set out in the Belize case and discussed above. Since Yam Seng, however, the Supreme Court has explicitly retreated from this approach, reasserting that, for a term to be implied in fact, the following criteria must all be satisfied:

(1) It must be reasonable and equitable; (2) it must be necessary to give business efficacy to the contract, such that no term will be implied if the contract is effective without it; (3) it must be so obvious that ‘it goes without saying’; (4) it must be capable of clear expression; (5) it must not contradict any express terms of the contract.^[30]

Despite Leggatt LJ’s assertion to the contrary in the recent case of Nehayan v Kent,^[31] satisfying these criteria in respect of a duty to perform contracts in good faith presents clear problems: in Nehayan itself, for example, Leggatt LJ expressly observed it to be “perhaps impossible to attempt to spell out an exhaustive description” of what the duty involves^[32] – thereby demonstrating that it cannot comply with the fourth criterion, that a duty “must be capable of clear expression”. Conversely, in Shell v Lostock,^[33] the Court of Appeal refused to imply a term in the comparatively precise wording that “[the Claimant] would not abnormally discriminate against [the Defendant] in favour of neighbouring and competing garages so as to render [its] petrol sales uneconomical” due to “the vagueness and ambiguity inherent in such words as ‘discriminate’ and ‘abnormality’”.^[34] In order to satisfy the criterion of clarity, therefore, it is necessary to specify precisely to which “types” of conduct the term applies. This would likely mean the implication of a number of lesser, more specific terms, equivalent in effect to the broader, “umbrella” duty of good faith. The same conclusion can be reached via the application of the “business efficacy” test:^[35] if one or more lesser, more specific terms would be capable of giving the contract such business efficacy as must have been intended by the parties, the greater can by no means be considered “necessary”.

In light of Marks & Spencer v BNP Paribas, therefore, the duty of good faith contract performance implied in fact can, at best, be regarded as an organising principle underpinning a number of lesser, more specific duties, equivalent in effect to the “broader” duty identified by Leggatt J. In Yam Seng itself, it would be trivial to achieve the same result via the implication of a number of lesser terms: the contractual relationship required a sufficient degree of cooperation in its performance that it was “necessary” to imply a term preventing the parties from making any false statements of fact upon the other was bound to rely in the performance of the contract. If no such duty was implied, a consistent pattern of dishonest conduct could never give rise to a right to terminate^[36] – even if the effect of the deception is sufficiently serious to rob the injured party of substantially the whole benefit that the parties intended him to gain from the contract.^[37] Similarly, in Yam Seng, the failure to positively disclose that ITC’s distributor in the domestic market of Singapore firstly refused to raise its prices, and then agreed to do so only with the addendum that it would take a further two months, had an equivalent effect to directly deceiving the Claimant: compare, for example, Shankland & Co v Robinson & Co^[38] and Notts Patent Brick & Tile Co v Butler,^[39] in which duties to positively disclose information were found to operate even in the (inherently adversarial) precontractual position.^[40]

Leggatt J is therefore justified in his finding that a duty of good faith contract performance, enforced via the mechanism of implied terms, might well (in certain circumstances) extend beyond a duty not to positively deceive the other party regarding performance of the contract.^[41] The extent to which it will do so is, however, highly sensitive to context: in JH Ritchie v Lloyds,^[42] for example, a duty was implied to disclose information even beyond the correction of omissions that are analogous to a direct deception, due to the precise relationship created between the parties by the factual background of the contract.^[43] Equally, in certain contexts, terms might be implied preventing “improper, commercially unacceptable or unconscionable” conduct (which Leggatt J also considered to fall within “the core requirement of honesty”):^[44] for example, Bristol Groundschool v Intelligent Data Capture^[45] and D&G Cars v Essex Police Authority^[46] – applying Yam Seng – both involved the implication of terms preventing conduct which, although directly facilitated by the existence of the contracts, did not accord with how the parties must reasonably have intended the contracts to function.^[47] The use of implied terms to regulate such conduct must, however, be limited to conduct relating in some way to how the parties intended the contract to function, or the criteria for the implication itself cannot be satisfied: in Chelsfield Advisers v QDREIC,^[48] for example, the Court refused to extend the principle from contracts of employment that a duty of good faith implied in law might extend to conduct unrelated to the contract^[49] beyond its limited context.

Leggatt LJ’s subsequent judgment in Nehayan v Kent^[50] can be regarded as an application of these principles. The alleged “bad faith” conduct in that case consisted of: a party to a joint venture contract extracting payments from his co-venturer by threatening to exploit his position as majority shareholder to block a deal upon which the survival of the venture depended; and, the same party deliberately concealing from his co-venturer that, whilst his co-venturer negotiated that deal, he was simultaneously conducting enquiries with the same third party with a view to disposing of his interest in the venture, which his co-venturer had expected to gain. As in D&G Cars and Bristol Groundschool, it was only by virtue of the contract’s existence that the Claimant was empowered to engage in this conduct – and in the case of the former conduct, it is clear that the parties cannot have intended that they should be free to exploit the contract’s existence to extort payments from one another. A term preventing economic duress is therefore “necessary” in much the same way as a term preventing deceit was, in Yam Seng. In the case of the latter conduct, however, one might well venture that freedom of contract requires that parties to such an agreement should be entitled to dispose of their interest in the venture howsoever they see fit – thus rendering a term restricting such disposal as neither “necessary” nor “obvious”. The precise factual background of the arrangement in this case, however, suggests otherwise: the parties in Nehayan – although sophisticated commercial actors – chose to enter into an oral contract, devoid of detailed express terms providing for the minutiae of their business operations. Such an arrangement is antithetical to the situation in Philips Electronique v British Sky Broadcasting,^[51] in which a lengthy and detailed contract was indicative that an omission to provide for the scenario at hand was deliberate – rather, the deliberate choice to enter into an oral contract suggests that, objectively, the parties had expectations of trust and loyalty far beyond the meagre demands of the express terms of the contract^[52] (this is emphasised by the parties repeatedly referred to one another as “friend” and “brother” throughout performance of the contract). Thus, the implication of a term preventing the parties from engaging in such “double dealing”^[53] was necessary in order to give effect to those expectations.

In each of the examples discussed, the specific “type” of bad faith conduct at issue was quite capable of control via the traditional tests for the implication of terms in fact – and, indeed, in the case of Nehayan, deciding the case by reference to a broader duty carries the danger of infringing upon the freedom of contract of parties to joint venture contracts where the same expectations of loyalty are not present.^[54] One might, therefore, question the necessity (and, indeed, the wisdom) of recognising any broader duty of good faith at all – the existing law of implied terms serving perfectly well as a “piecemeal solution” to protect the reasonable expectations of the parties, without need for recourse to broader notions of good faith and fair dealing. Nonetheless, the failure to acknowledge that good faith is an organising principle which underpins the development of such “piecemeal solutions” has led to certain inconsistencies in the law’s development, which could be reconciled by the recognition of such a principle; this is well illustrated by the implication of terms regulating one particular “type” of bad faith conduct: the misuse of a contractual discretion.

IV. Controlling Contractual Discretion

Since long before the time of Yam Seng, it has been recognised that a discretion expressly conferred by a contract might be subject to control. These controls have recently been summarised by the High Court as being that such a discretion “must be exercised in good faith, for the purpose for which it was conferred, and must not be exercised arbitrarily, capriciously or unreasonably”:^[55] for example, in Government of the Republic of Spain v North of England Steamship Co,^[56] the discretion was not exercised in any meaningful sense at all; and, in Equitable Life Assurance Co v Hyman^[57], the exercise of a discretion to reduce annual bonuses in line with a drop in the annuity rate in a contract under which a guaranteed annuity rate was payable clearly frustrated the “purpose” for which the discretion was granted. In each of these cases, the implication of a term to control the discretion was “necessary” in order to enable the contract to function as the parties must have intended – in that, by expressly granting a discretion for a specified purpose, the parties must have intended that the discretion was exercised for, and only for, this purpose.

The notion that a discretion must not be exercised arbitrarily, capriciously or unreasonably echoes the test for irrationality in the context of judicial review, from Associated Provincial Picture House v Wednesbury Corporation;^[58] and, although the courts have – in cases such as The Product Star (No 2)^[59] and Lymington Marina v MacNamara^[60] – cautioned against the import of public law principles, their preferred wording that the discretion must not be exercised “wholly unreasonably” clearly imposes a considerably more permissive standard than mere reasonableness,^[61] and is functionally equivalent to the Wednesbury test. This justifies the courts, in cases such as Braganza v BP Shipping^[62] and Watson v Watchfinder,^[63] in overlooking such concerns by referring directly to the Wednesbury test in this context. Applying the tests for the implication of terms to this restriction, it can firstly be observed that mere reasonableness would quite obviously be insufficiently clear; nor would it be “obvious” that the parties would agree to such a subjective term, or “necessary” to enable the contract to function as the parties must have intended. The stricter standard, however, sets a sufficiently high standard of “unreasonableness” that any justifiable “reason” for the discretion’s exercise is permissible: for example, Lomas v JFB Firth Rixson^[64] demonstrates that acting in one’s own commercial selfishness is perfectly acceptable – which, by a lesser standard of unreasonableness, might not be the case. The restriction is, therefore, sufficiently certain; and, furthermore, “so obvious that it goes without saying”, and “necessary” to enable the contract to function as the parties must, objectively, have intended: if the parties had intended to act so arbitrarily and capriciously as to defy all reason, they should have expressed it.^[65]

The final component of the restriction – that a discretion must be exercised “in good faith” – does not seem to add anything to this: although some authorities (for example, Rix LJ’s comments in Mallone v BPB Industries^[66]) suggest that, in theory, it should – as “someone may act irrationally whilst being honest” – this appears to be based on a subjective definition of good faith, and is therefore incompatible with the purely objective approach to the implication of terms in fact.^[67] One might well recall, at this point, Lord Greene’s paraphrasing of the Wednesbury test that the threshold is a decision that is “so unreasonable that it might almost be described as being done in bad faith”;^[68] the control of a contractual discretion is, therefore, essentially a test to ascertain whether the discretion should objectively be considered to have been exercised in bad faith – rendering bad faith a consequence, rather than a prerequisite, of its satisfaction.

Clearly, therefore – although encompassing these principles – Leggatt J’s approach to good faith does not, at this point, add anything substantive to the regulation of good faith contract performance in the context of a contractual discretion. One might still, therefore, question the wisdom of recognising any broader principle of good faith. Somewhat ironically, the answer to this can be found in subsequent authorities’ attempts to back away from the decision in Yam Seng: in Mid Essex Hospital Services NHS Trust v Compass Group,^[69] for example – consciously distancing itself from Yam Seng, by finding an ongoing service agreement for the provision of cleaning and catering services within NHS hospitals not to constitute a “relational” contract – a distinction was drawn between “a simple decision whether to exercise an absolute contractual right” and “making an assessment or choosing from a range of options”:^[70] only in the latter case will a term be implied controlling the exercise of discretion. Thus, the Trust’s express contractual right to decide whether or not to award “service failure points” in respect of shortcomings in the services provided was unconstrained by any implied term: having decided to award points, the amount of points to award was clearly defined by the express terms of the contract, meaning that the discretion amounted to “a simple decision whether to exercise an absolute contractual right”.^[71] Portsmouth City Council v Ensign Highways^[72] – which, similarly, related to the discretion to award service failure points in respect of shortcomings in a PFI contract – was therefore distinguished on the basis that, in that case, the Council had the discretion to decide how many such points to award.^[73] This distinction between “binary” and “complex” discretion^[74] was approved by the Court in Greenclose v National Westminster Bank^[75] and explicitly set out in such terms in Myers v Kestrel Acquisitions.^[76] There are, however, numerous difficulties with such an approach.

Firstly, it is unsupported by the existing body of case law: McKay v Centurion Credit Resources^[77] and Lymington Marina v MacNamara,^[78] for example, each involved the implication of terms controlling a “binary” discretion^[79] – and The Product Star (No 2),^[80] which was specifically referred to in the judgment as an example of a “complex” discretion,^[81] in fact simply related to a binary decision as to whether or not to dock in a given port. Conversely, although there is little authority to suggest that a “complex” discretion might be exercisable unconstrained by any implied term, there is no reason in principle why this could not be the case: for example, if a contract between a homeowner and a decorator provides that the homeowner should be free to determine the colour scheme, there is no reason why full effect should not be given to his decision – no matter how arbitrary, capricious or unreasonable his choice may be.

Secondly, formulating the test in terms of whether or not the discretion is “binary” – as the Court did in Myers v Kestrel Acquisitions^[82] – gives rise to an absurdity: there can be no difference in principle whether a contract grants a discretion to choose between two, three or a dozen options. This concern is somewhat alleviated by noting that Jackson LJ did not appear to have in mind that the exercise of any choice between two options should be unconstrained: rather, it is a choice between two specific options – “whether or not to exercise an absolute contractual right”^[83] – that constitute a “binary” discretion in this sense. This second objection nonetheless illustrates a potential pitfall in adopting Jackson LJ’s approach – and, furthermore, it must be noted that a large number of interrelated “binary” discretions could be indistinguishable in effect from a “complex” discretion, thus rendering the distinction entirely arbitrary.

Thirdly – and most troublingly – the distinction is not justified by reference to any matter of policy, nor by reference to the properly applicable tests for the implication of terms in fact. As noted above, these tests will usually be satisfied with regard to a term controlling the exercise of a contractual discretion. The only exception is where, on a proper construction of the contract, it appears that the parties intended for the discretion to be completely unconstrained: for example, in cases such as Hamsard v Boots^[84] and Monde Petroleum v Westernzagros,^[85] the courts have found an unqualified power to terminate a contract as being completely unconstrained, as preventing the parties from exercising such a discretion would be analogous to compelling them to enter into a new contract – which cannot, objectively, have been their intention.^[86] The Court therefore erred, in Socimer International Bank v Standard Bank London,^[87] in finding that even a discretion granted in “sole and absolute terms”^[88] was automatically limited by an implied term not to be exercised arbitrarily, capriciously or unreasonably.^[89] It may, however, have been fully entitled to reach the same outcome by reference to the principle in Spain v North of England:^[90] the discretion, in Socimer, empowered one party to determine the value of certain assets. If such a value were set arbitrarily, or in bad faith, the discretion cannot properly be said to have been exercised at all – and certainly not for the purpose for which it was conferred – meaning that its control would remain justified.^[91]

The Court, in Mid Essex Hospital Services, was therefore unjustified in deciding the case by reference to the discretion being a “binary” choice, rather than to the objective intention of the parties. Although it will often be so that a “binary” discretion (such as an option to terminate) must be intended to be unconstrained, whereas a broader discretion must be limited in some way, this is not necessarily the case, and regarding it as so risks departing from the form and purpose of the implication process. The “rule” that a “complex” discretion will be subject to control whereas a “binary” choice will not, must, therefore, be regarded as a factual trend – and not as a generalisable rule of law. By application of the proper test, the Court, in Mid Essex Hospital Services, would have been quite justified in reaching the same result: the contract in that case made detailed provision for the award of service failure points – and, in fact, included a limited express duty of good faith which did not apply to the circumstances at hand. In such a case, it should clearly be presumed that the omission of any control on the discretion should be deliberate – see, for example, Phillips Electronique v British Sky Broadcasting.^[92]

This branch of case law, Sheppard notes, “[relies] on the contextual specificities of contractual meaning”^[93] in order to limit Yam Seng’s applicability to particular commercial contexts. As has been discussed, this is wholly unnecessary: if, in such a context, the parties must reasonably have expected that each would pursue their own interests – unfettered by any implied duties – then the proper application of Yam Seng would allow them to do so. There is, furthermore, a certain irony to this approach: despite objections to Yam Seng largely stemming from the need for commercial certainty^[94] and the desire to give effect to the will of the parties (as embodied by the express terms of the contract),^[95] attempts to limit its impact have the effect of producing inconsistent, unpredictable and arbitrary rules. The recognition of a general, but highly contextual duty of good faith – as set out in Yam Seng – as an organising principle behind this area of law would, whilst allowing for the same outcome in each case, alleviate these shortcomings in the law.

V. Controlling Legal Discretion

A further inconsistency that would be reconciled by the recognition of good faith as an organising principle can be found in in MSC Mediterranean Shipping Company v Cottonex Anstalt,^[96] in which – without referring to his earlier judgment, in Yam Seng – Leggatt J attempted to bring the rule in White & Carter (Councils) v McGregor,^[97] that a choice to affirm a contract in response to a repudiatory breach might not be exercisable if the injured party lacks a “legitimate interest” in continued performance, into line with the above authorities controlling a contractual discretion. Although no guidance was provided in White & Carter itself as to what constitutes a “legitimate interest” in this context save that “the de minimis principle [should] apply”,^[98] a number of authorities build on precisely what is meant by the term – a detailed discussion of these is provided by Cooke J’s judgment in The Aquafaith,^[99] to which, in MSC v Cottonex, Leggatt J defers.^[100]

Although the analysis of this discussion is, “perhaps inevitably, inconclusive”,^[101] a number of general principles can be discerned. Firstly, as a matter of practicality, the contract clearly cannot be affirmed if the innocent party’s continued performance would require some degree of cooperation from the repudiating party,^[102] in which case a remedy should only be available in damages – although note that, in many cases, simply making goods available for delivery or being available to perform services is sufficient performance, even if the other party refuses acceptance.^[103] Secondly, the burden is on the party in breach to demonstrate that the innocent party has no legitimate interest in performing the contract, and this burden is not discharged simply by showing that the benefit to the other party is smaller than the loss to the contract breaker.^[104] Thirdly, the innocent party can only be compelled to accept the repudiation if damages would be an adequate remedy^[105] – this will not be the case if, for example, they have suffered no pecuniary loss.^[106] Finally – and, for our purpose, most significantly – the innocent party can be compelled to accept the repudiation if (and only if) “an election to keep the promise alive would be wholly unreasonable.”^[107] For the purpose of this final test, a distinction must be drawn between mere reasonableness and a decision which is “wholly unreasonable” (which, indeed, is the same phrasing applied by Arden LJ in Lymington Marina v MacNamara,^[108] in the context of contractual discretion): if it were sufficient that a court simply considered the decision to affirm the contract to be unreasonable, as Lord Reid noted in White & Carter, the rule would “introduce a novel equitable doctrine that a party was not to be held to his contract unless the court in a given instance thought it reasonable so to do.”^[109]

In MSC v Cottonex, after setting out these principles, Leggatt J went on to note the “increasing recognition in the common law world of the need for good faith in contractual dealings”^[110] – and, in particular, the line of authority regarding the implication of terms restricting contractual discretion, as discussed above. Without going into any detail as to the extent of the duty to exercise a discretion in good faith (in either context), he concluded that “[it] should [not] make any difference in principle” whether an election constitutes “the exercise of discretionary powers conferred by the express terms of the contract … [or] the choice whether or not to terminate the contract in response to a repudiatory breach”;^[111] thus, he regarded each line of authority “as concerned with materially identical questions and as establishing essentially the same test.”^[112]

Despite Leggatt J’s assertion to the contrary,^[113] there are, in fact, key conceptual differences between the control of a discretion arising as a matter of law and one expressly conferred by the parties: primarily, that legal rules do not directly arises as a result of the parties’ intentions – from which any good faith duty implied in fact is derived. A term implied in law, for example, arises not by virtue of the parties’ intentions but “as a necessary incident of a definable category of contractual relationship”.^[114] Thus, terms implied in law may be more difficult to exclude than those implied in fact^[115] – and, certain terms implied by statute are incapable of exclusion.^[116] Other examples of situations in which rules of law may overwrite the expressed will of the parties include the common law’s treatment of various particular “types” of contract term (such as penalty clauses, entire agreement clauses, “agreement to agree” clauses and certain applications of limitation clauses), the “red hand rule” and the plethora of statutes that now exist regulating unfair contract terms and consumer rights.

It must, however, be noted that few, if any, of these rules of law can truly be divorced from the intention of the parties^[117] (or, at times, the principle of good faith):^[118] many are a simple matter of construction, which must be derived from the intention of the parties;^[119] others simply operate as default rules, to which reasonable persons will be presumed to assent unless they expressly provide otherwise;^[120] yet more address the situation in which one party may not have assented to the term if they had knowledge of it,^[121] or sufficient bargaining power to object.^[122] There is, therefore, a certain logic to treating rules such as these as equivalent to the terms of the contract for the purpose of ascertaining whether they may be subject to control. By this logic, it is immaterial whether the matter at issue is a rule of law acting to control the expressed will of the parties (such as an express term being rendered unenforceable)^[123] or, as in the instant case, the will of the parties (as embodied by an implied term) operating to control the exercise of a power arising as a matter of law.

This suggests that Leggatt J is quite justified in treating a legal discretion as equivalent to one arising under the express terms of the contract; however, as has been discussed, such a discretion will only be subject to control if the parties did not intend for the discretion to confer an absolute and unqualified right.^[124] A choice arising as a matter of law is clearly analogous to such a right: in the absence of clear and unambiguous wording to the contrary, the parties must, objectively, have considered that any rights arising as a matter of law would be unfettered. Comparison can be drawn to the construction of terms restricting liability for negligence or deceit, in which even clearly worded provisions whose natural meaning would encompass such liability may be unable to restrict the operation of rules of law, unless it is spelled out in explicit terms that it should do so.^[125] Viewed through the lens of good faith, however, these examples are fully consistent with restricting a legal discretion by reference to such a principle: in the absence of clear wording to the contrary, parties acting in good faith will presume that they should not intend to act dishonestly or negligently; likewise, they might well presume that the other should not seek to enforce or discharge their contractual rights and obligations in bad faith, where they lack a legitimate interest in doing so. Clearly, therefore – as in the context of controlling contractual discretion – good faith constitutes a useful organising principle by which otherwise contradictory rules of law can be reconciled. Leggatt J is therefore justified in his view that the rule in White & Carter and the control of a contractual discretion are “concerned with materially identical questions”,^[126] meaning that, in principle, there is no reason to distinguish between them; there are, however, subtle differences in the application of each control.

As previously noted, the use of the phrase “wholly unreasonable”^[127] in the control of legal discretion echoes that used in the context of implied terms restricting contractual discretion.^[128] As in that context, therefore, it may seem appropriate once again to import the concept of Wednesbury unreasonableness from administrative law: a decision can only be said to be “wholly” unreasonable if it is “so unreasonable that no reasonable [decision maker] could ever have come to it”.^[129] As applied, however, the test seems to differ somewhat in its application: The Puerto Buitrago,^[130] for example – which represents a rare example of the rule being successfully pleased – established that, in determining whether a decision is wholly unreasonable, account must be taken of the burden placed upon the party in breach by keeping the contract alive; in the control of a contractual discretion, by contrast, acting purely out of one’s own commercial selfishness may well be permissible.^[131] This decision, furthermore, appears to contradict the general rule that a decision to affirm the contract is not unreasonable simply because “the benefit to the [innocent] party is small in comparison to the loss to the contract breaker”^[132] – and, therefore, calls into question the precise threshold of unreasonableness required.

Conversely, the rule in White & Carter is, in other respects, less onerous than the control of contractual discretion: in addition to the extra criteria that must be satisfied in order for a choice to be controlled, such a decision may only be challenged if it is “wholly unreasonable”^[133] – by contrast to the comparatively broad controls applicable to a contractual discretion, which “must be exercised in good faith for the purpose for which it was conferred, and must not be exercised arbitrarily, capriciously or unreasonably.”^[134] The rule is further limited in that, rather than requiring simply that the decision of whether or not to terminate the contract should be made in good faith (howsoever ascertained), it seems to apply only to the circumstance in which a person unreasonably insists upon performance of the contract – there is no authority to suggest that it might equally prevent a person from unreasonably terminating the contract in response to a repudiatory breach. Thus, in Arcos v Ronaasen,^[135] a buyer of barrel staves was entitled to terminate the contract in bad faith – in order to escape from a bad bargain (which, ordinarily, is anathema to contract law) – despite having suffered no loss in respect of the breach, simply because the term breached was classified as a “condition”.

Whilst it is apparent that Leggatt J was, therefore, mistaken in his conclusion that the control of contractual discretion and the rule in White & Carter “[establish] essentially the same test”,^[136] there seems to be little reason in principle why a modest incremental step should not be taken in placing them on an equivalent footing. As discussed, the law would be quite justified in treating legal rights as equivalent to contractual rights in this context; and, furthermore, extending the treatment of contractual discretion to this situation would add clarity and consistency to the rule in White & Carter: Hooley, for example, notes that “[adopting] a similar approach to the exercise of a right of termination for breach as … to the exercise of a contractual discretion … would probably catch a case like Arcos v Ronaasen”,^[137] thereby bringing it into line with White & Carter. Whilst it is often said that the rigidity of the principle in Arcos brings with it a desirable degree of certainty,^[138] this overlooks that the classification of a term as a “condition” can, at times, be somewhat ambiguous^[139] – and the growing use of the so-called “innominate terms”,^[140] which may or may not give rise to a right to terminate,^[141] equally calls into question the logic of preserving an inconsistent rule simply for the sake of certainty. Furthermore, as discussed, the applicability and extent of the rule in White & Carter remain somewhat ambiguous; there is, therefore, no small irony in the refusal to recognise the rule as an emanation of an implied duty of good faith – thereby bringing it into line with the control of a contractual discretion – solely by reference to legal certainty.

VI. A Reconciliatory Increment

As outlined above, the parallel evolution of equivalent rules, in equivalent circumstances, has given rise to numerous inconsistencies. This can be regarded as symptomatic of the “piecemeal approach” to good faith characteristic of the law of England and Wales: where numerous rules are created, independently of one another but in analogous contexts, some inconsistencies are bound to arise in the natural course of the law’s development. As previously discussed, these inconsistencies are averted by deriving a duty to act in good faith from the intention of the parties – whether via implied terms or construction – which may be flexible enough to ensure that such a duty does not go beyond the level of individualism envisaged by the parties. Furthermore – and in addition to reconciling the aforementioned inconsistencies – Leggatt J’s attempt, in MSC v Cottonex, to extend the principles governing contractual discretion to a discretion arising at law is fully consistent with the incremental approach favoured by the law of England and Wales (it being a relatively small step to hold that the same rules should govern broadly similar tests arising in equivalent circumstances), thereby clarifying and illuminating the already ambiguous doctrine of White & Carter.

It is, therefore, regrettable that – on appeal^[142] – Moore-Bick LJ did not embrace Leggatt J’s approach to MSC v Cottonex (although he was able to instead decide the case by reference to frustration, meaning that it remains open to a bolder Court of Appeal to take this modest incremental step), instead setting out a number of reasons why he did not consider recognising a duty of good faith to be “necessary or desirable”.^[143] In particular, he noted: that the courts have time and again refused to recognise any general duty to act in good faith;^[144] that the law of England and Wales prefers to develop “piecemeal solutions in response to demonstrated problems of unfairness”,^[145] rather than by reference to broad, overarching principles; and, “that if a general principle of good faith were established it [might] be invoked as often to undermine as to support the terms in which the parties have reached agreement.”^[146] Each of these objections, however, presume a general, universal standard of good faith – meaning that they are inapplicable to the highly contextual approach to good faith, derived from the intention of the parties, set out in Yam Seng and advocated herein. Thus, whilst a general, universal duty (as rightly rejected by the Court of Appeal in Mid Essex Hospital Services v Compass Group,^[147] to which Moore-Bick LJ referred in setting out his first objection) would indeed be incompatible with the incremental, piecemeal approach of the law of England and Wales – and might well “be invoked as often to undermine as to support the terms in which the parties have reached agreement”^[148] – a contextual duty, derived from the intention of the parties could never be invoked to do so: rather, the primary purpose of such a duty is to ensure compliance with the agreement itself, as objectively intended by the parties (that is, fidelity to the bargain).^[149] Furthermore, Moore-Bick LJ himself noted that “it is well recognised that broad concepts of fair dealing may be reflected in the court’s response to questions of construction and implication”;^[150] thus, recognising good faith as an organising principle – which, adding nothing substantive to the law, would constitute a relatively modest incremental advancement – adds doctrinal consistency to those solutions, leading to more principled and predictable law. One such example of these benefits is evident from the doctrinal context of MSC v Cottonex itself: although Moore-Bick LJ appears to have been correct in deciding that the contract in that case had been frustrated,^[151] extending the principles governing the control of a contractual discretion to a legal discretion – as Leggatt J tried to do at first instance – would, as has been discussed, reconcile a number of inconsistencies within this latter area.

Similarly – as outlined in detail above – certain inconsistencies are particularly evident within the courts’ treatment of implied terms controlling contractual discretion.^[152] Although this article has attempted to establish some general order behind these disparities, such reasoning is not evident within the authorities themselves: rather, the decisive factor often seems to be whether a court embraces the recognition of a duty of good faith – as in Portsmouth CC v Ensign Highways^[153] – or, as in Mid Essex Hospital Services v Compass Group,^[154] finds some trivial reason to reject it in order to reach the desired outcome.^[155] As previously discussed, in each of these cases the courts would have been entitled to reach the same outcome by reference to the principle in Yam Seng; recognising good faith as an organising principle, not extending but explaining the existing rules, would, therefore, enable the courts to reach a fair outcome in cases such as these without need for recourse to such judicial distortions.^[156]

VII. Further Increments in Canadian Law: Bhasin v Hrynew

In his approach to MSC v Cottonex, Leggatt J drew inspiration from the “increasing recognition in the common law world of the need for good faith in contractual dealings.”^[157] In particular, he drew attention to “the unanimous judgment of the Supreme Court of Canada in Bhasin v Hrynew”^[158] – which, it must be noted, likewise drew inspiration from his earlier judgment in Yam Seng – “holding that good faith contractual performance is a general organising principle of the common law of contract which underpins and informs more specific rules and doctrines.”^[159] As has been discussed, the recognition of good faith as an organising principle, by which otherwise contradictory rules can be reconciled, adds much-needed coherence to the law of implied terms, improving the clarity, consistency and predictability of the law – and constitutes a fairly modest incremental step in the law’s development. The Court, in Bhasin, however, went further than mere recognition of a pre-existing principle, also applying a second incremental advancement, somewhat less modest than the first: “to recognise … a common law duty which applies to all contracts to act honestly in the performance of contractual obligations.”^[160] Given the clear advantages of the former of these proposed advancements – and the apparent contradictions between the latter and the existing body of case law^[161] – the validity of this reasoning, and its potential applicability to the law of England and Wales, therefore warrants close scrutiny.

Bhasin itself related to a contract analogous to a franchise agreement, under which Mr Bhasin, acting as an agent, marketed education saving plans on behalf of Canadian American Finance Corp (“Can-Am”). The contract was to run for three-year terms, automatically renewing after each such interval subject to a “non-renewal” clause, exercisable on six months’ notice. A regulator subsequently raised concerns with Can-Am regarding certain compliance issues and asked it to appoint an officer to review its agents’ compliance; Can-Am duly appointed Mr Hrynew – another agent, who traded in direct competition with Mr Hrynew and was attempting to undertake a “hostile takeover” of Mr Bhasin’s business – to this role. Mr Bhasin objected to allowing Mr Hrynew to view his confidential records. Can-Am responded with a number of false statements (namely, that Mr Hrynew was subject to confidentiality and segregation of activities duties, and that the regulator had refused a request to allow an outside officer to conduct the reviews) in order to persuade Mr Bhasin to comply; and, furthermore, failed to disclose when specifically asked by Mr Bhasin that its future plans included restructuring its business such that Mr Hrynew seized control of Mr Bhasin’s business. When Mr Bhasin still refused to comply, Can-Am exercised the non-renewal clause, bringing the contract to an end. This severely damaged the value of Mr Bhasin’s business – and, he averred, constituted a breach of an implied duty of good faith.

After conducting a detailed survey of good faith in contract performance in Anglo-Canadian law^[162] – including reference to Yam Seng v ITC^[163] – the Court, perhaps unsurprisingly, cited with approval the conclusions of the Ontario Law Reform Commission that the common law’s stubborn refusal to recognise a general doctrine of good faith has produced “an unsettled and incoherent body of law” that has developed “piecemeal” and which “is difficult to analyse.”^[164] This approach, Cromwell J asserted, “produces results that are not consistent with the reasonable expectations of commercial parties.”^[165] Thus, the Court concluded that it was appropriate to take the two incremental steps identified at the outset:^[166]

“The first step is to acknowledge that good faith in contractual performance is a general organising principle of the common law of contract which underpins and informs the various rules in which the common law, in various situations and types of relationships, recognises obligations of good faith contractual performance. The second is to recognise, as a further manifestation of this organising principle of good faith, that there is a common law duty which applies to all contracts to act honestly in the performance of contractual obligations.”

The first of these increments, as noted, would be equally justified within England and Wales, and would bring clarity, cohesion and doctrinal consistency to the law. The second, however, is a far greater leap: as previously noted, the applicable tests for the implication of terms in fact do not justify the implication of a broader, universal duty of good faith (as proposed in Bhasin, and consistently rejected by the courts of England and Wales),^[167] but rather, a number of lesser, more specific terms, based on the individual contextual background of a particular contract. The Court, in Bhasin, bypasses this problem by finding that “a new duty of honest performance … should not be thought of as an implied term, but a general doctrine of contract law that imposes as a contractual duty a minimum standard of honest contractual performance”, and which “operates irrespective of the intentions of the parties.”^[168]

As has been noted – and made emphatically clear by the courts^[169] – this development of the law by reference to broad, overarching principles is incompatible with the preferred mechanisms of the law of England and Wales; thus, the duty imposed in Bhasin is best analysed by reference to the implied duty of good faith discussed within this article. Terms implied in fact, however – as discussed – must necessarily be derived from the reasonable intentions of the parties. It may, therefore, seem that a term implied in law would be the more appropriate analogy;^[170] this is further illustrated by the Court, in Bhasin, justifying this second incremental step by reference to the existence of terms implied in law imposing a duty of good faith in certain circumstances.^[171] The implication of terms in law, however, requires that the contract fall within a definable type,^[172] rather than simply providing for universal rules of law.^[173] There is, therefore, no comparable analogue by which a general standard of good faith, operating irrespective of the parties’ intentions, can be imposed;^[174] however, as shall become clear, the precise nature and extent of the duty proposed in Bhasin is, in many respects, comparable to Leggatt J’s approach in Yam Seng – accordingly, for present purposes, it shall be analysed by reference to this principle.

A striking similarity between Bhasin and Yam Seng is the extent of the “minimum standard of honest contractual performance”^[175] identified in the former case, which constitutes “a simple requirement not to lie or mislead the other party about one’s contractual performance.”^[176] This echoes Leggatt J’s sentiment that “in any situation it is dishonest to deceive another person by making a statement of fact intending that other person to rely on it while knowing the statement to be untrue”^[177] – and is, furthermore, reminiscent of the “lesser” implied term proposed above, that a person should not make false statements upon which the other is bound to rely in the performance of the contract. As noted, this term should properly be implied into any contract in which performance might depend on the provision of information; and, as it can have no application in any other contractual environment, it is immaterial whether this is a general duty implied into all contracts or a specific one applying only to contracts in which it is capable of producing an effect. The only difference in practice between the minimum standards applied by each approach, therefore, is that parties are expressly unable to exclude the “minimum standard” imposed by Bhasin. As a breach of this standard would necessarily entail commission of the tort of deceit, however, the parties remain unable to exclude such a term unless they specify their intent to do so in very clear terms – and, it is hard to envisage the circumstance in which reasonable parties, entering into a contract characterised by a high degree of cooperation would do so. For present purposes, therefore, it is permissible to consider the core requirement of honesty in Bhasin and Yam Seng as being fully equivalent in their operation.

Beyond this “minimum standard”, Cromwell J notes that “the precise content of honest performance will vary with context and the parties should be free in some contexts to relax the requirements of the doctrine so long as they respect its minimum core requirements.”^[178] As with the core requirement itself, this extension of the duty is barely distinguishable from the highly contextual view of good faith set out in Yam Seng and advocated herein. Although little else is provided by way of guidance as to how such a duty is ascertained, Cromwell J went on to note:

Recognising a duty of honesty in contractual performance poses no risk to commercial certainty in the law of contract. A reasonable commercial person would expect, at least, that the other party to a contract would not be dishonest about his or her performance. The duty is also clear and easy to apply.^[179]

These assertions betray further similarities with the implication of terms in fact: in each case, any duty will be determined by reference to the precise contractual context at issue; each can be derived from the reasonable expectations of commercial persons; and, each must, by definition, be “clear and easy to apply”. One might well venture that this latter assertion warrants far greater justification: although Cromwell J highlights that, in the absence of a recognised doctrine of good faith, there is still a great deal of uncertainty in this area,^[180] this by no means justifies introducing a general, overriding duty whose extent and limits remain unknown.^[181] Such concerns can only be alleviated by embracing the derivation of the duty entirely from the reasonable expectations of commercial persons in the precise contractual context at issue.

If the general duty of good faith performance in all contracts is to be internally consistent, it must, therefore be regarded as functionally equivalent to the duties implied in fact under the principle in Yam Seng. Applying the principle in Yam Seng to the facts of Bhasin, the outcome of the case would be clear: a contract characterised by such a high degree of cooperation and predictable performance necessarily requires that parties should neither directly mislead one another with regard to its performance, nor fail to disclose information upon request when such failure would have an equivalent effect to a direct deception; the consistent pattern of dishonest conduct by Can-Am would, therefore, breach the implied duty of good faith, in much the same way as the consistent pattern of dishonesty by ITC in Yam Seng itself. The exercise of the non-renewal clause, however, clearly constitutes the exercise of an absolute contractual right: the Court in Bhasin acknowledged that “the parties did not intend or presume a perpetual contract, as they contracted that either party could unilaterally cause it to expire on any third anniversary”.^[182] Thus, notwithstanding its prior dishonesty, Can-Am should have been fully entitled to end the contract: this will generally be the case with regard to any termination clause^[183] – and, a non-renewal clause essentially serving to regulate the circumstances in which a further contract will be concluded, interfering with its exercise would constitute a gross violation of freedom of contract.

On this latter point, however, the Court in Bhasin reached an entirely different conclusion: finding the pattern of deception to be “intimately connected to Can-Am’s … exercise of the non-renewal provision”,^[184] it concluded that the termination of the contract itself breached the duty of good faith, awarding damages in respect of the consequent loss in value of Mr Bhasin’s business. For the reasons given, this is unjustified: Can-Am should have been quite entitled to insist that, if Mr Bhasin wished to enter into a new contract past the three year fixed term, he must agree to allow Hrynew to review its records.^[185] If, in entering into such a contract, the Claimant relied upon the falsehoods provided by Can-Am, he might well have had a remedy in breach of contract or misrepresentation; but freedom of contract requires that a person should not be compelled to enter into a contract when to do so would be contrary to their own interests. Thus, Cromwell J’s reassurance that the duty of good faith applied in Bhasin does not interfere the with freedom of contract^[186] is an entirely hollow claim.

Overreaching the duty of good faith in such a fashion somewhat justifies concerns that a duty of good faith may be too unpredictable,^[187] and could represent an incursion upon freedom of contract.^[188] These dangers, however, arise from circumstances which are not applicable to the law of England and Wales: as noted, the second of Cromwell J’s increments is incompatible with the “piecemeal approach” of English law;^[189] and, furthermore cannot properly be said to constitute a “modest increment”^[190] as suggested, but is rather a somewhat grand leap. Given the wealth of authorities providing, in very clear terms, that such an approach is incompatible with the law of England and Wales,^[191] one would be hard pressed to imagine an English judge taking the same step.

However, given the contradictions that have arisen in the post-Yam Seng case law, there is nonetheless a danger that the doctrine could expand in unforeseeable ways, producing comparably aberrant outcomes to that in Bhasin – see, for example, the arbitrary and unjustified distinction between “binary” and “complex” discretion in cases such as Myers v Kestrel Acquisitions.^[192] Such a danger does not, however, justify departing entirely from the principle in Yam Seng – which, as discussed, on its proper construction adds much-needed clarity and consistency to this area of law. The recognition of Leggatt J’s view of good faith as not substantively changing the law but constituting an organising principle behind the implication of terms in fact, as advocated within this article, would preserve these benefits whilst abating the doctrine’s potential expansion – much akin to how the Court in Marks & Spencer v BNP Paribas^[193] treated Lord Hoffman’s approach to implied terms in AG of Belize v Belize Telecom^[194] – thereby rooting the doctrine firmly within the confines of its theoretical foundation whilst allowing the fruits of doctrinal consistency to flourish.