I. Introduction
A contract is a risk allocation instrument that predicts and regulates the future relations between the parties. Often, over the course of time, parties agree to act differently in certain respects from that which is stipulated in the formal contract, or in practice their conduct deviates from the contract. Such cases result from tension between two aspects of the contractual relationship. On the one hand, at the stage of the formation of the contract, there is a desire to fully regulate the relations between the parties, or at least to require formalisation of future updates of the contract. To this end, many contracts contain a clause stipulating that deviations from the contract will be binding only if they are made in writing and formally. These are referred to as no oral modification (NOM) clauses. On the other hand, because contractual relationships are rich and dynamic, many believe that even when the original contract contains an NOM clause, focusing on the original contract and disregarding later agreements and practices thwarts the wishes of the parties and leads to inequitable results.
This tension is reflected in the different attitude of the law of various countries regarding the enforcement of NOM clauses. According to one position, such clauses should not be enforced. This position was recognised in the second Restatement in the US,Footnote 1 and has been accepted in several states there,Footnote 2 as well as in Australia,Footnote 3 and certain European countries.Footnote 4 Until recently, this position guided significant cases in the UK.Footnote 5 The main justification of this approach, both in case law and in academic literature,Footnote 6 is the desire to honour the autonomy of the parties and their absolute freedom to modify their prior agreements, including agreements on the manner by which the contract should be modified. A clear expression of this position was given by Justice Cardozo in the Alfred C Beatty v Guggenheim Exploration Company judgment. In his words: “Those who make a contract, may unmake it. The clause which forbids a change, may be changed like any other. The prohibition of oral waiver, may itself be waived.”Footnote 7 At least some of those advocating this position nonetheless suggest that a NOM clause should not be ignored entirely, but that stronger evidence of the intention of the parties to modify the contract is required than if the clause did not exist.Footnote 8
According to the competing position, NOM clauses are valid. This is the prevalent view, although usually subject to exceptions, in the United Nations Convention on Contracts for the International Sale of Goods (CISG),Footnote 9 in international legislative initiatives in Europe,Footnote 10 in some European countries,Footnote 11 in the Uniform Commercial Code in the US,Footnote 12 and even in some states in the US, the most prominent of which is New York.Footnote 13 Recently, in MWB Business Exchange v Rock Advertising,Footnote 14 this view was adopted by majority opinion (in the Supreme Court) as the guiding position in the UK.
With respect to the value of autonomy, this position clashes with the previous one in two respects. First, on the ethical level, it holds that the proper realisation of the value of autonomy does not lie in the recognition of the parties’ right to modify a contract they have signed, but rather in the parties’ self-enforcing ability regarding the manner in which the contract is modified.Footnote 15 Second, proponents of this position argue that even at the factual level, conduct or oral agreement that deviates from that which is written in the contract can reflect a desire for legal change in relation to the issue in question, but does not necessarily reflect a desire to abolish the NOM.Footnote 16 From time to time, even representatives of this position admit that it is often not equitable to demand that a party that has relied on the promise or conduct of the other party that has deviated from the written agreement be nevertheless bound by the original contract.Footnote 17 It has been argued, however, that these concerns can be addressed by the doctrine of estoppel,Footnote 18 emphasising that the law must ensure that adherence to this doctrine is limited and specific, so that in most instances the legal validity of the NOM clause is preserved.Footnote 19
In this article, we wish to add a significant layer to the theoretical discussion of NOM clauses, and subsequently, to propose a new model for their practical regulation. In our opinion, the debate regarding the validity of NOMs reflects a fundamental tension underlying modern contract law between the formal written contract and the contractual relationship as a whole. Due to the work of scholars such as Macaulay,Footnote 20 Macneil,Footnote 21 Scott,Footnote 22 Collins,Footnote 23 Campbell,Footnote 24 Kimel,Footnote 25 Zamir,Footnote 26 BrownswordFootnote 27 and other contract law theorists, there has been growing recognition since the late twentieth century that the formal contract does not exhaust the relationship between the parties to it, and that there are significant aspects in relationships that are not expressed in the formal contract. Despite such recognition, however, there is still serious controversy over whether contract law should give contractual validity to the entire set of relationships and expectations between the parties, without material distinction between the aspects expressed in the formal contract and those that are part of the relationship but not formally enshrined in the contract. Two main approaches have emerged on this issue. One, held by scholars of relational contract theory, seeks to give contractual validity to all the understandings, both formal and informal, which make up the relationship between the parties.Footnote 28 By contrast, neo-formalist scholars acknowledge the existence of rich informal components in the relationships between the parties,Footnote 29 but hold that contract law should avoid giving contractual validity to these components, which should be relegated to the domain of extralegal incentive systems.
The controversy over the contractual status of the informal dimensions of relationships, which include, for example, social and cultural aspects, is broad and encompasses diverse components of contract law. The best-known implications of this controversy relate to contract formation and, in particular, to the requirement of indefiniteness,Footnote 30 interpretationFootnote 31 and frustration.Footnote 32 This article demonstrates how the controversy between neo-formalist and relational contract theory may contribute to the theoretical analysis and practical regulation of NOM clauses.Footnote 33
Our theoretical analysis contributes to the current debate concerning NOM clauses in three areas. First, from the perspective of the parties’ autonomy, the article seeks to offer a deeper understanding of the wishes of the parties, distinguishing between intention focused on the legal relationship and intention focused on extra-contractual relations. Second, our analysis exposes the fact that the regulation of NOM not only reflects the intention and relationship of the parties but that it also constitutes and shapes it. Therefore, our analysis goes beyond a “party autonomy” perspective and explains how legal enforcement of NOM may influence the parties’ behaviour and relations. Third, the article adds an element of economic analysis that clarifies the roles played by efficiency and institutional considerations, such as the need for certainty and reducing the costs of litigation.
Following a theoretical analysis, the article proposes a comprehensive model for regulating NOM clauses.
The key innovation advanced by the model is a shift from dichotomous regulation to context-dependent regulation that includes two main aspects.
First, we distinguish between three types of relationship. The first type is between two sophisticated and equally powerful parties, concerning which neo-formalist arguments appear to be more applicable. Therefore, in such relationships, it is appropriate to enforce NOM clauses. The second is a relationship between two unsophisticated parties of equal power, with respect to whom the arguments advanced by relational theory, which do not favour enforcing NOM clauses, are more persuasive. The third type of relationship is characterised by power disparities. In such circumstances, when the deviation from the formal contract favours the stronger party, it is appropriate to enforce NOMs, but in situations in which the deviation from the formal contract favours the weaker side, as a rule, we recommend not enforcing the NOM clause.
Second, we offer criteria to distinguish between cases in which the deviation from the contract indicates the parties’ desire for legal changes to their relationship from cases in which the deviation should not be granted contractual validity. Our criteria include: the parties’ awareness of the deviation from the formal agreement; the duration, consistency and significance of the modification; the degree of investment in contract formation, and the formality of the ongoing relationship between the parties; and the reasons behind the original deviation from the formal contract and the current demand to return to it.
The contextual model introduced in this article does not present a one-dimensional position regarding each relationship, but recommends ways to refine conclusions and take into account the opposing considerations in each case. On the one hand, according to the proposed model, even when the NOM clause is held to be valid, the results can still be refined by the doctrine of estoppel. On the other hand, according to our approach, even when the relational theory is compelling, and NOM clauses ought not to be granted full legal validity, the neo-formalist arguments are not neglected and gain expression through auxiliary tests aimed at ensuring that the modification made by the parties reflects their intention to make legal changes as well.
The controversy between the relational and the neo-formalist approaches, and its contribution to the theoretical debate on the question of enforcement of NOM clauses, are reviewed in Part II of this article. Part III offers an innovative and nuanced model for practical regulation of NOM clauses.
II. A New Look at NOM Clauses
The relational/neo-formalist controversy spans three main dimensions:
(1) The autonomy dimension seeks to shape the law in a way that reflects the parties’ intention. Therefore, it involves sociological arguments on how to understand parties’ legal intention;
(2) The constitutive normative dimension focuses on the influence of the law on the parties’ behaviour and relations;
(3) Finally, the institutional and efficiency dimension addresses the interplay between the costs of drafting, litigation and certainty, both for the parties and for the legal system as a whole.Footnote 34
Accordingly, below, we analyse how the controversy between the relational and the neo-formalist approaches in these three dimensions contributes to the debate regarding NOM clauses.
A. A New Set of Autonomy Arguments
1. The Preference of Relations over a Written Contract
Classic contract lawFootnote 35 has been criticised over the years for focusing exclusively on the formal interaction between the parties to the contract (the so-called “paper deal”), ignoring many aspects of the relationship (the so-called “real deal”) that do not gain expression in the formal contract.Footnote 36 This type of criticism was expressed most systematically by relational contract theory,Footnote 37 which argues that the relationship between the parties to a contract is much richer than the prescriptions formally captured by the contract. The sociological aspect of the theory is empirically based on studies showing that contracting parties ascribe great importance to extralegal considerations, such as maintaining a good reputation with the public at large, a desire for future transactions, moral perceptions, mutual trust, their standing in the relevant community, and more.Footnote 38 Relational contract theory has developed beyond this sociological insight, and, with time, has become a normative theory arguing that contract law should be shaped in a way that takes account of the parties’ actual conduct, and that the focus should not be limited to the formal interactions between them.
Relational contract theory provides theoretical depth for the approaches that oppose enforcing NOM clauses. According to this theory, the formal contract is intended primarily to build trust between the parties. However, as relations develop and trust evolves, the importance of the initial formal agreement diminishes over time.Footnote 39 Hence, the parties’ conduct over the course of performance is the best indicator of their actual intentions and of the modifications to which they have agreed.Footnote 40 Therefore, precisely out of respect for the value of autonomy, contractual meaning must be ascribed to the evolving relations between the parties and the informal agreements formed between them over time.Footnote 41 It is therefore clear that this approach endorses the position that rejects NOM clauses, and recognises that a contractual agreement cannot be set in stone in its early stages.
2. Neo-formalist Support for NOM Clauses
Neo-formalist theory emerged in opposition to relational contract theory. In contrast to the classic formalist approach, which was blamed for lack of awareness of the informal aspects of contractual relations, the neo-formalist approach takes account of these elements. Yet, awareness of the existence and importance of informal aspects of contractual relations notwithstanding, neo-formalists argue that contractual validity should be granted only to the formal contract, not to the parties’ broader relations, which are not reflected in the formal contract. According to neo-formalist sociological analysis, even when the relationship between the parties is rich, and includes non-formal aspects, the parties still prefer that legal regulation focus strictly on the formal aspect of this relationship. It has been arguedFootnote 42 that contracting parties wish to separate the extralegal norms that characterise the parties’ relationship in peacetime (that is, as long as the relationship lasts), from the legal norms that apply in wartime (that is, when a conflict arises between the parties and the relations are no longer expected to continue).Footnote 43 While parties’ behaviour in peacetime frequently deviates from the formal contract, the parties themselves want the legal arrangement of contractual relations during any war to reflect the written contractual arrangement and not the practices that evolved during peacetime.Footnote 44
The common justifications for granting validity to NOM clauses have been based on the right of the parties to bind themselves against future modifications of the contract.Footnote 45 The neo-formalist approach clarifies that by including such a clause, the parties clearly delineate the distinction between arrangements that apply in wartime, from those that apply in peacetime. Therefore, enforcing NOMs does not merely reflect blind compliance to the parties’ pre-commitment, but also a respect for the distinction between wartime and peacetime relationships.
3. A New Intermediate Position
In existing legal discourse, in addition to the two extreme positions analysed by Lord Sumption in MWB v Rock, which sweepingly reject or support the possibility of the parties binding themselves through NOM clauses, there is the intermediate position expressed by Lord Briggs. This position recognises the parties’ right to deviate from the NOM clause, but holds that even if the conduct or statement that deviates from the stipulations in the contract reflects a desire to move away from the contractual arrangement regarding the particular point in question, it does not necessarily reflect an intent to change the NOM clause.Footnote 46 In the spirit of this intermediate position, some legal systems have required explicit reference to both a desire to change the arrangement, and a desire to change the NOM clause, as a condition for deviating from a NOM clause.Footnote 47
In the previous section we showed that the distinction between wartime and peacetime led neo-formalists to support enforcement of NOM Clauses. In contrast, in this section, on the basis of the same distinction, we offer a new version of an intermediate position. According to this new intermediate position, the distinction between wartime and peacetime suggests that, at times, conduct-based and oral deviation from the written arrangement not only fails to indicate a desire to change the NOM clause, as Lord Briggs suggests, but often, it is not possible to infer from it any desire to make a permanent legal change that will be applicable in times of war. Therefore, according to our position, even in cases where there is no NOM clause, and all the more so in instances where such a stipulation does exist, one must ensure that the conduct deviating from the contract reflects an agreement to affect a permanent legal modification of the formal contract. Only after it is clear that such consent exists, should one also examine, in the spirit of Lord Briggs's admonition, whether the parties intended to rescind the NOM clause.
The proposed intermediate position contributes to the debate in another way. Relational contract theory and the neo-formalist approach offer opposing positions with respect to enforcement of NOM clauses. In contrast, the intermediate position proposed by Lord Briggs expects the judge to decide, ad hoc, what the parties’ intention was in the instant case, but does not guide the judge on how to achieve that goal. The position we offer adopts the intermediate position to the extent that it acknowledges that there is no one sweeping answer to the question of whether or not to enforce a NOM clause. However, in the third Part, we will suggest distinctions between different relationships, and auxiliary tests which distinguish different circumstances. The purpose of these is to guide judges on when the clause should be enforced.
B. Beyond Autonomy: The Normative Aspects of the Debate
The debate between relational contract theory and neo-formalism is not limited to the sociological argument regarding how to understand the parties’ intention. In many instances, there is in addition a normative value-laden argument. In other cases, economic and institutional factors also play a part in determining whether to enforce NOM clauses.
1. Relational Theory Supports the Non-enforcement of NOM Clauses
Relational contract theory emerges in two versions – communitarian and libertarian, each presenting different views on the extent to which the theory should be implemented.Footnote 48
The libertarian version of relational contract theory seeks to grant contractual validity to norms that reflect values such as interpersonal solidarity, cooperation, and mutual consideration,Footnote 49 but the inclusion of non-formal value-based norms is limited to those that have been adopted in practice (even if not formally) by the parties by their conduct, or by virtue of their shared culture. This version refuses to impose moral values and considerations of justice that the parties did not intend to apply to their relations.Footnote 50
For the purposes of the discussion in the previous section, which focused on the autonomy claim, the libertarian version of relational contract theory sufficed to justify the non-enforcement of NOM clauses. However, there is also a communitarian version of relational contract theory. The communitarian version imposes moral values and considerations of justice on contractual relationships, even in situations in which these values and considerations do not reflect the parties’ actual behaviour and intentions.Footnote 51
Applying the communitarian version of relational contract theoryFootnote 52 to the discussion of NOM clauses provides two important justifications for non-enforcement of NOM clauses that go beyond the autonomy justification. First, it leads to the conclusion that after the parties have already deviated from the original contract by their own conduct, and especially when one of the parties has already relied on the change, insistence on observing the formal requirement leads to unjust results.Footnote 53 Therefore, it is necessary to apply equitable principles to justify deviation from the NOM clause.Footnote 54 Second, granting legal validity to the informal components of the relationship, strengthens those components. Legal involvement helps create new social domains, and expands the relations between the parties. For example, it has been argued, that the publication of legal sanctions in specific cases helps disseminate information about the parties’ misconduct, thus enabling market players to create and consolidate extralegal sanctions and incentives where these are lacking or insufficiently developed.Footnote 55 By contrast, rigid separation between the paper deal and the real deal, and focusing merely on the former, may create an incentive for opportunistic behaviour, thereby damaging the trust between the parties as well as the relationship between them.Footnote 56 Given that enforcement of NOM clauses means that the law is to ignore the informal elements in the parties’ relations, it is reasonable to assume that such a legal regime may prevent the development of these elements in the long term. Thus, enforcement of NOM clauses should be opposed not only because honouring the autonomy of parties to a contract also means honouring their ability to change their minds, but also because of broad considerations that seek to enrich contractual relations and strengthen their informal aspects.
2. The Neo-formalist Normative Arguments in Favour of Enforcement of NOM Clauses
While the sociological argument assumes that the law should reflect the intention of parties, the normative argument deals with the manner in which contractual relations and parties’ expectations should be constructed in the first place. According to the neo-formalist normative argument, non-judicial dimensions and extralegal norms improve the relations between the parties and reduce, ab initio, the likelihood of disputes going to court. By contrast, juridicising the complete relationship between the parties, especially the informal aspects, is liable to lead to the extinction of informal aspects, harming the relations between the parties. For example, if the parties to the contract know that concessions made over the course of the life of the contract will create an obligation for them, they will avoid making such concessions, and adopt a more rigid approach towards the other parties, which will harm their overall relationship.Footnote 57 According to neo-formalist theory, it is precisely the legal focus on the written contract, and the distinction between the legal and non-legal dimensions of the relations between the parties, that will improve the parties’ relations and provide a “safe zone” for them in the long term, where they can demonstrate generosity and friendship without such actions being used against them on a legal level.Footnote 58 This neo-formalist position cautions against granting legal validity to developments in the parties’ relations after the formal contract has been signed, even when the original contract did not include a NOM clause. This is all the more true when the parties themselves have sought to create a buffer between the written legal contract, and future non-legal developments by means of the NOM clause. Social interest in enabling the creation of a non-legal private trust domain supports the validation of such clauses.
C. Institutional and Efficiency Considerations
Even in the absence of NOM clauses, neo-formalist scholars and their relational contract peers differ on the issue of giving effect to conduct by the parties that deviates from the written agreement. This debate mixes institutional considerations with efficiency concerns.
Neo-formalists assume that, for the most part, formal agreements are clearer than conduct or verbal utterances, and that focusing on the formal written contract encourages stability and prevents litigation. Therefore, the neo-formalist approach rejects validation of informal agreements.Footnote 59 Moreover, even when there is a dispute with respect to the interpretation of the formal contract, the costs of a legal argument that focuses on the linguistic content of a formal stipulation in a contract are lower than those of legal arguments that demand proof of conduct or verbal statements that followed the contract's execution. For this reason, focusing on the formal content would, of necessity, reduce litigation costs.Footnote 60 Therefore, according to the neo-formalistic approach, later contractual modification that is not made formally should not be given effect.Footnote 61
Adherents of relational contract theory do not reject the stability and certainty arguments. Nevertheless, according to the relational approach, the recognition of informal aspects encourages contractual certainty, because often it is later conduct that clarifies the parties’ intention.Footnote 62 Moreover, recognising future behaviour as an enforceable modification saves contract formation costs, as it is impossible to address every future scenario in a contract.Footnote 63 Thus, scholars of the relational contract theory argue that efficiency considerations support providing legal validity to parties’ behaviour that deviates from the formal contract.
In the instances in which the contract itself does not contain a NOM clause, we believe that both positions – neo-formalist and relational – are partially correct in this debate. Alongside cases in which focusing on the written contract contributes to certainty, prevents litigation and saves costs, there are other cases in which the opposite is true. Because of this complexity, we should allow the parties, who best know the concrete circumstances of the deal they are planning, to determine in advance whether they prefer litigation based on the written contract, and therefore prefer to invest in contract formation to reduce later costs, or whether they prefer to grant legal validity to informal understandings to promote certainty and reduce costs.
This analysis has significant implications on the proper legal attitude to enforcement of NOM clauses. Enforcing NOM clauses allowing the parties to choose between litigation limited to the written contract, and litigation that requires reference to subsequent unwritten developments, thereby signals which arrangement is more efficient. Thus, honouring the NOM clause reflects not only respect for the parties’ autonomy, but also helps realise the institutional goal of promoting certainty and efficiency.Footnote 64
At first glance, therefore, it appears that institutional concerns and efficiency arguments unequivocally support the neo-formalist position to enforce NOM clauses.Footnote 65 This conclusion however, is based on the assumption that the existence of the clause indicates that the parties have considered the dilemma between the costs of drafting and the costs of later litigation, and preferred to save the latter.
As demonstrated below, this conclusion must be qualified for relationships between equal and sophisticated parties, when it is reasonable to believe that the inclusion of the NOM clause, or alternatively, its waiver, reflects the parties’ conscious and informed intent regarding the types of cost they prefer.
In this context, enforcing NOM clauses echoes conventional English commercial law, according to which a signed document should be treated as definitive of the terms of the contract.Footnote 66
By contrast, in the case of non-sophisticated parties, or where there are power disparities between them, we doubt that this analysis holds. In the next section, we will deal with the need to distinguish between these three prototypical relationships.
III. A New Model for Regulating NOM Clauses and Modifying Contracts by Conduct
Based on the new theoretical understanding offered in this article, we propose a coherent outline for the legal regulation of NOM clauses, and for modification of contracts by conduct, even absent a NOM clause.
A. Contextuality
For a long time, relational and neo-formalist theories were situated at two opposite poles of the theoretical discourse in contract law. Recently, however, writers have concluded that the disagreement between the relational and neo-formalist theories was largely due to the fact that each theory considered concrete relationships, but at the same time, sought to formulate arrangements that would apply to contract law in its entirety.Footnote 67 Hence, scholars belonging to both groups have come to the realisation that the scope of the controversy is not large, and that instead of arguing across the board, the focus should be on adapting each approach to the relationships relevant to it.Footnote 68 In this section, we apply the contextual approach to NOM clauses, and propose a detailed legal model explaining how the legal attitude toward NOM clauses should be affected by different types of relationships.Footnote 69
1. Adopting the Neo-formalist Approach: Sophisticated Parties and Organisations
In academic literature pertaining to contract law,Footnote 70 and in certain respects also in case law,Footnote 71 the recognition of commercial relations between sophisticated and legally advised parties as a distinct category of contracts, with respect to which neo-formalist norms should be adopted, is gaining traction. In the context of the current article, these norms call for stricter tests for recognising contract modifications by conduct, even without NOM clauses, and grant contractual validity to NOM clauses in which the parties explicitly reject the possibility of modifying the contract by conduct.
To date, the concept “sophisticated parties” has not been sufficiently defined in case law, despite the use of the term.Footnote 72 Schwartz and Scott have suggested a definition of sophisticated parties with reference to the size of the firm, its type of incorporation and its fields of occupation.Footnote 73 However, we fear that a technical definition, no matter how brilliant, will fail to fully realise the rationale for which we seek to refer to sophisticated parties as a distinct category for the purposes of NOM clauses, and contractual modification by conduct. Therefore, instead of a dichotomous approach, and a one-dimensional, technical definition, we wish to propose several variables that would define parties as sophisticated for the purposes of enforcing NOM clauses. These variables include: (1) legal advice accompanying not only the stage of the drafting but also the life of the contract; (2) the commercial and professional experience of the active players, which in many instances are repeat players in the market; and (3) businesses with a complex organisational structure, which distinguishes managerial parties from agents operating in the field. When these variables exist, the logic underlying the neo-formalist case to enforce NOM clauses and grant limited validity to behavioural modifications not made in a formal agreement even in the absence of such a clause, is stronger.
In the case of experienced and legally advised parties, one can reasonably assume that the fact that the parties have failed to formally anchor the changes is not a result of absentmindedness, and is not attributable to a lack of legal representation. Therefore, one can assume that the parties themselves have not viewed these changes as legally binding.
When dealing with repeat players, and in particular with large entities, the extralegal mechanism described by neo-formalist literature as deterring breach of promises, such as harm to one's reputation, may be crucial. Thus, there is a certain logic to the parties leaving some aspects of their relationship as non-legal, in reliance on these mechanisms.
In the case of commercial relationships between economically-oriented parties, efficacy considerations that seek to enable the parties to strengthen contractual certainty, and to reduce future litigation costs, are valid.Footnote 74
Finally, our support for adopting the neo-formalist approach is particularly strong in the case of large organisations, where those who formulate the contracts, and those who carry them out, are not the same individuals. In these cases, a clause denying the possibility of modifying the contract by conduct is in practice a managerial and organisational tool, intended to prevent agents of the organisation from making irreversible changes.Footnote 75 Note that in cases in which public entities such as the state are involved, an administrative dimension is added to the clauses that negates the modification of contracts by conduct, preventing the possibility of officials in the field from granting benefits without authority (e.g. deviating from tender procedures or protocol).Footnote 76
On this point, we wish to add several comments.
First, despite our attempt to define sophisticated parties, we recognise that this term, in fact, is not fixed, but can be determined along a spectrum, and that grey areas exist. This fact alone may impair legal certainty. We will discuss in detail the tension between considerations of certainty and the need to formulate a complex model that provides tailored solutions to a wide range of circumstances in the paper's concluding section.
Second, together with distinguishing the various prototypes of parties to a contract, this paper also proposes a series of tests to aid in identifying them at the level of the concrete deal; for example, how detailed the original contract was, or to what degree the relationship between the parties in question was characterised by formality. These tests, to a degree, relax the binary nature of the distinctions between the types of contractual parties, and are particularly significant in the grey areas where classification of the parties’ relationship is challenging.
Finally, despite our support for the adoption of the neo-formalist approach with respect to sophisticated parties and large-scale organisations, in some circumstance, the results of a NOM clause must be blunted by the application of equitable doctrines, such as estoppel. In the next section, we discuss the application of the doctrine of estoppel, and suggest legal mechanisms for ensuring that it does not entirely undermine the NOM clause.Footnote 77
2. Adopting the Relational Contract Approach with Respect to Unsophisticated, Evenly Matched Parties
In the case of unsophisticated parties of equal bargaining power, we tend to recognise the benefits of the relational approach, which does not honour NOM clauses.Footnote 78 Like the definition of sophisticated parties, the definition of unsophisticated parties is not binary or dichotomous. It refers to: (1) small entities or private individuals; and (2) parties that lack continuous access to legal advice, and therefore are represented, at most, during the stage of the contract's drafting and execution. Under such circumstances, the case for non-enforcement of NOM clauses is strengthened for the following reasons.
First, parties that lack ongoing representation over the life of the contract may not be fully aware of the need to formally draft the changes, and thus their choice not to formally write down the modification should not be interpreted as reflecting their wishes not to be legally bound by them.Footnote 79
Second, from an economic perspective, requiring a formal definition of every contractual change, necessitating the hiring of a lawyer, is not cost effective in such a relationship.Footnote 80
Third, whereas in large organisations, NOM clauses are sometimes justified as a means for management to limit the agents operating in the field, in small organisations or in the case of private individuals who are not incorporated, this consideration is less relevant (since the parties shaping the deal and those carrying it out may be identical). The overlap between the personal identity of the management shaping the deal and the agents operating in the field carrying it out blurs the distinction between the stage of shaping the deal – which represents, according to the neo-formalist claim, the rules for wartime – and the stage of the ongoing relationship – which represents, according to the neo-formalist claim, the rules for peacetime.
In light of the totality of the abovementioned reasons, in this type of relationship we recommend continuing the policy reflected by the Court of Appeal in MWB v Rock,Footnote 81 according to which a change of conduct can be viewed as a modification of the NOM clause, in the setting of the parties’ freedom to deviate from former undertakings.
Even in this regard, however, we must remember our caution that not every deviation by behaviour from written contractual instructions indicates a desire to change the contract permanently in a legally binding manner for the future as well. According to our analysis, even in the case of evenly matched unsophisticated parties, only in instances in which the court is persuaded that deviation from the written contract constitutes a desire to modify it permanently should the NOM clause not be enforced. By contrast, when there is concern that the deviation from the contract does not reflect a desire for a permanent and binding legal change, the written contract must be adhered to, in line with, but regardless of, the NOM. Below we propose auxiliary tests to help distinguish between the two types of cases.Footnote 82
Thus far, we have discussed cases in which the justification for non-enforcement of NOM clauses as offered by relational contract theory offered is based on the liberal argument that granting effect to the parties’ later conduct, rather to their earlier formal agreement, more closely reflects the parties’ true intentions, thus respecting their autonomy. However, as we have seen, the communitarian version of relational contract theory is willing at times to justify non-enforcement of NOM clauses for reasons of fairness and integrity. In this spirit, we believe that, in the extreme cases in which exercising the NOM clause and disregarding the behavioural modification would result in extreme injustice and significant departure from the parties’ legitimate expectations as these have evolved over time, NOM clauses should not be enforced. In such cases, one should consider applying doctrines such as material unconscionability to cancel the NOM clause.Footnote 83 Alternatively, when business–to-business contracts are involved, one should consider the application of the Unfair Contract Terms Act 1977 (UCTA), which prevents a party from relying on unreasonable provisions which “render a contractual performance substantially different from that which is reasonably expected of him”Footnote 84 regarding NOM clauses.
3. The Complex Case of Power Disparities
A third type of relationship is characterised by power disparities. In the case of such relations, it is necessary to distinguish whether the deviation serves the stronger or the weaker party.
When the deviation from the written contract works in favour of the stronger party, it makes sense to give effect to the NOM, because the stronger and legally represented party can be expected to insist that the agreed deviation by conduct be given formal legal expression. Moreover, in such cases, it is not at all clear that the weaker party is aware of the deviation from the contractual agreement, or its binding legal significance.
In contrast, in the case of a deviation from the contract that favours the weaker party, our position is more complex. In this case, over the life of the contract, the parties have deviated from the written contract in favour of the weaker party, and subsequently, the stronger party seeks to renounce the deviation and re-assert its rights under the original contract. Normally, power disparities are precisely the situations in which stronger parties make sure to incorporate into the contract a clause that prohibits modification by conduct, and therefore the stronger party intends to rely on this clause. As a rule, we believe that in these cases, the position of non-compliance with NOMs is justified, and the considerations expounded by the relational contract approach, especially its communitarian version, are pertinent.Footnote 85
For this reason, in such cases, when dealing with a consumer contract within the meaning of that term in the Consumer Rights Act,Footnote 86 a NOM clause would be very likely to “cause a significant imbalance in the parties’ rights and obligations under the contract to the detriment of the consumer”.Footnote 87 Hence it is considered to be an “unfair term of a consumer contract”, and thus is not binding on the consumer.Footnote 88 However, we believe that even in cases that are not formally classified as consumer contracts, but where there is still considerable disparity between the experience of the parties, or their access to legal advice,Footnote 89 there is scope for applying general doctrines that encourage interference with contracts, such as unconscionability to facilitate rescission of the NOM clause.Footnote 90
At the same time, there are two significant exceptions to our general position on this point.
First, we accept the right of complex organisations to deny their agents operating in the field the authority to amend legal contracts without the supervision of the authorised bodies within the organisation. Therefore, when the NOM clause serves as a managerial means of controlling decision-making in the organisation, it should be respected.Footnote 91
At the same time, precisely in such cases, it is also necessary to protect weaker parties who have relied on promises and representations by the organisation or its officers. Therefore, for these types of cases, middle-ground solutions should be developed. While generally recognising that a NOM clause negates contract modification by conduct, exceptions should be made for justice and fairness considerations (in which case, the NOM may be nullified or granted only partial validity). As shown below, informed use of equitable doctrines such as estoppel can result in the acceptance of such middle-ground solutions.
Second, in the long term, the ability of stronger parties to rely on NOMs may at times benefit weaker parties, whereas not honouring these clauses out of a motivation to protect the weaker parties, may work to their detriment. The motivation lies in the neo-formalist distinction between peacetime and wartime rules. A stronger party that knows that a concession in favour of a weaker party will not be binding in the future because of the presence of a NOM clause, may allow itself to assume a generous stance toward the weaker party. In contrast, an approach that negates the validity of the NOM clause, may prevent the stronger party from showing generosity toward the weaker party and making a concession in a particular case, for fear that such voluntary conduct may become binding. Therefore, even if the NOM clause is not enforced, caution must be exercised not to interpret every deviation from the contract of a stronger party in favour of a weaker one as reflecting a willingness to make a legal change. This warning is especially true when the change in conduct is not mutual (involving an exchange of benefits) but is a unilateral gesture on the part of the stronger party.
B. When Does the Conduct of the Parties Indicate Agreement to Legal Change?
Our analysis reveals that in the case of unsophisticated parties or those characterised by power disparities, when the deviations are in favour of the weak party, the neo-formalist justifications for enforcing the NOM clauses are weakened. It appears, therefore, that in the context of these relationships, binding contractual validity must be given to the actual conduct of the parties, even when it is contrary to the formal provisions of the contract, and even if the contract includes a NOM clause.
In the present section we wish to refine this conclusion. In too many cases, the discussion of NOM clauses implicitly assumes that by having deviated by conduct from the provisions of the contract, the parties intended to legally modify these provisions, and therefore the discussion focuses on whether this legal modification is possible in view of the NOM clause. This approach, however, ignores the claims of the neo-formalist position, according to which, in light of the distinction made by the parties between the wartime and peacetime relations, it is not always possible to deduce from a deviation from the provisions of the contract a desire to change the relations between the parties in a permanent, legal way.Footnote 92 This neo-formalist caveat should be taken seriously even without a NOM clause,Footnote 93 and certainly when such a clause exists. In this section, we propose several auxiliary tests aimed at distinguishing between cases in which the deviation from the contract indicates a desire to modify the contract, and cases in which the deviation should not be granted contractual validity. Our proposed auxiliary tests combine an in-depth evaluation of the parties’ intent with normative and institutional considerations.
1. The Parties’ Awareness of the Deviation from the Formal Agreement
Conventional wisdom views modification of a contract as the formation of a new contract, and therefore examines the validity of the modification based on the usual tests for contract formation. By contrast, especially in the case of unsophisticated parties, there are instances in which a party is not at all aware that its conduct has deviated from the contract. Apparently, according to the objective test for contract formation, subjective lack of awareness makes no difference. But in the spirit of the neo-formalist position, we believe that unless it has been proven that the parties were aware that their actual behaviour deviated from the formal agreement, such deviation did not indicate consent to legal change. This is because, the parties trust their lawyers to draft the legal arrangement in writing, with the understanding that their future conduct will not necessarily have legal validity. Therefore, it is unreasonable to assume that they have checked whether their conduct conforms to the written contract. Under these circumstances, examining whether the parties have objectively deviated from the contract as is common in the case of contract formation, is not sufficient. It is necessary to verify that the parties have deliberately decided to deviate from the contract to provide legal validity for this deviation.
Nevertheless, this conclusion must be qualified because the awareness test is not only factual but also normative. Therefore, in some cases, it is appropriate to apply a concept of constructive knowledge and to attribute such awareness to the party that should have known about the modification by virtue of professional status and education, even if knowledge in practice remains unproven.
2. Duration and Consistency
We propose to balance or bridge the neo-formalist and relational approaches by differentiating between one-time conduct or a sporadic deviation in the short term on the one hand, and consistent deviation in the long term on the other. Only a consistent deviation over a long time should be recognised in our context as having legal validity. The neo-formalist approach is usually correct in cautioning against granting contractual validity to modification by conduct in cases of one-time conduct, both based on in-depth analysis of the parties’ intent and on an economic consideration that seeks to enable the parties to show consideration for each other and act generously when required, without such conduct being held against them. By contrast, in the case of unsophisticated parties, and when dealing with constant change over time, the relational approach is better suited to the actual reality in arguing that ignoring the change and demanding a return to the written contract, which clearly does not represent the relationship in practice, is unjustified and does not reflect the understanding of the parties.
Relational contract theory is also correct in stating that an informal agreement to deviate from the formal contract is formed during the course of the relations between the parties, without a concrete point in time in which the understanding between the parties has taken shape. Therefore, flexible tests considering the relations between the parties as a whole are needed (i.e. tests that can identify agreements between parties that have evolved linearly over time, with or without specifying a concrete point in time at which the contract was modified).Footnote 94
3. Significance of the Modification
Another way of expressing the concern for recognising contractual modification by conduct is to test its significance. According to this test, the court must recognise modifications by conduct only in matters that do not lie at the core of the contract. The demand that every minor change be made in writing is burdensome for unsophisticated and underrepresented parties. In contrast, in the case of material change to the core content of the contract, conduct should generally not be granted contractual validity, and certainly not contrary to a NOM clause. The expectation that a significant change be made in writing is justified from both from the point of view of the parties’ intention and from the economic perspective of cost savings.
Nevertheless, an exception to this principle is in order because, in some situations, conduct that deviates significantly from the provisions of the contract completely undermines the foundations of the contract and renders the insistence on the execution of the written contract irrelevant. Such cases do not constitute contract modification by conduct but rather a complete cancellation of the original agreement, which can also be accomplished both orally and by conduct.Footnote 95 Note, incidentally, that this exception is consistent with economic considerations that oppose legal validation of the parties’ conduct because of the difficulty of proving such conduct.Footnote 96 In the case of a dramatic change that completely undermines the contract, modification is easily proven.
4. Degree of Investment in Contract Formation and the Formality of the Ongoing Relationship between the Parties
Another way to bridge the dichotomous approaches is by taking into account the level of formality that characterises the relations between the parties. Parties that have invested time and money in formulating a detailed contract that considers a range of issues and scenarios that may materialise in the future, have signalled their intention to regulate their legal relations through the contract, and not to recognise conduct deviating from the contract as ground for its modification. By contrast, parties that choose not to invest in the drafting of the original contract and have not elaborated their agreement in detail, indicate the formation of a flexible framework, their intention to form dynamic relations, and the possibility of contract modification by conduct.Footnote 97
Beyond examining the level of formality of the parties at the beginning of the relations, we propose a cross-sectional examination of the parties’ conduct over the ongoing contractual relationship. According to this test, when over the course of the life of the contract the parties conducted themselves formally it is difficult to accept the claim that in a particular case the parties sought to change the original contract by conduct, without a formal modification. By contrast, when over their ongoing relations the parties did not insist that deviations from the original contract be formalised, they reinforce the position of the relational approach that seeks to recognise the conduct of the parties as a basis for their intention to legally change the contract.
Along with its focus on the parties’ intention, the proposed distinction between different levels of formality in the formulation of the contract and the course of the relations also has institutional-economic logic. Recall that one of the neo-formalist arguments against recognising contractual modification by conduct was the desire to give parties an incentive to formulate a detailed contract that can prevent future litigation.Footnote 98 A policy that assigns decisive weight to the level of detail of the contract gives an incentive to parties that wish to ensure that the court will act in accordance with the contract to invest in its drafting.Footnote 99 Similarly, the desire to encourage stability and certainty is likely to provide parties with an incentive to make sure that modifications are granted contractual validity and are carried out formally.
Nevertheless, there are other cases in which efficiency considerations discourage investment in the original drafting of the contract. For example, when it is difficult to anticipate the full range of future scenarios and circumstances, it may be wrong to give parties an incentive to invest in careful and costly drafting of the contract and in formal execution of any deviation, since the cost involved may exceed the deliberative cost of recognising the modification by conduct. Similarly, there are cases in which efficiency considerations discourage investment in the formal elaboration of the ongoing relationship. For example, in situations of material change that undermine the foundations of the original agreement, the cost of proving modification is not high, and at times is lower than the cost of drafting a new formal contract. In cases of this type, the law should adopt the relational contract approach, which recognises contract modification by conduct
Clearly, in the case of a significant change that is considered tantamount to the cancellation of the previous contract, there is no need to insist on a formal cancellation. By a similar argument to that in the previous section,Footnote 100 modification by conduct is sufficient.
5. The Reasons for the Original Deviation and for the Demand to Return to the Formal Contract
Taking into account normative considerations regarding the morality of ongoing contractual relations,Footnote 101 the reason behind deviation from the provisions of the contract may affect the question of the legal effect of the modification. When the modification is required because of the failure of one of the parties to fulfil the stipulations of the contract, this should not be considered consent to the modification. In contrast, if the modification is made necessary by objective circumstances, though such circumstances may not grant a party the right to demand the modification of the contract without the other's consent, it is more appropriate to grant contractual validity to unilateral modification by conduct, even if it has not been formally agreed.
Furthermore, a party's demand to return to the provisions of the original formal contract should also be examined from a moral-ethical perspective. To the extent that the demand to revert to the original contract results from the fact that prior circumstances allowing the deviation from the original contract no longer apply, the demand to revert to the written agreement is justified. Conversely, when both parties appear to have agreed to modification by conduct, and the reason for the demand to revert to the original contract stems from a desire to harass the other party or gain tactical benefits with respect to other disputes between parties, which are not relevant to the modification itself, the legitimacy of the demand to revert to the original contract diminishes.
Two reservations are relevant in this regard. First, especially in cases of power disparities, one of the parties may not have protested in real time to the modification due to fear of the other. In these situations, protest is legitimate when the relationship is over. Moreover, at times a party may have agreed to show restraint toward the modification by conduct so long as peaceful relations between the parties continued. In these cases, the party that did not protest in real time against the modification but raised the issue after the relationship deteriorated should not be blamed. Therefore, this test, which examines the reason for reversing the modification, must be applied with caution, showing sensitivity to the distinction between behaviour expected in peacetime and that which is expected in wartime.
A “reciprocity” test can provide another ethical distinction between those who wish to revert to the original contract as a tactical move and those who do so as part of a legitimate desire to return to wartime rules after the relations between the parties deteriorated. According to this test, to the extent that a system of mutual concessions has evolved between the parties, it is unfair for one party to withdraw from one of the concessions and demand a return to the written contract. By contrast, to the extent that concessions were unilateral, the legitimacy of the party that has renounced its rights as expressed in the original contract is strengthened.
C. The Role of the Doctrine of Estoppel when the NOM Clause Is Valid
The MWB Business Exchange v Rock Advertising Footnote 102 ruling opened the door to enforcement of NOM clauses. The discussion in this article supports the ruling in specific circumstances, including the case of evenly matched sophisticated parties. Nevertheless, our discussion has shown that even when enforcing the NOM clause, consideration should be given to the harm that may be caused to the party that has acted in reliance on the conduct or explicit statements of the other party or its representatives.Footnote 103 Even legal systems having an established tradition of enforcing NOMs, such as in some states in the US,Footnote 104 refine this practice through the use of the estoppel doctrine, which protects a party's reliance on the conduct of the other party in such circumstances.
The doctrine of equitable estoppelFootnote 105 requires the following conditions:
Where: (a) one person (the inducing party) plays a role in the adoption by another person (the relying party) of an assumption of fact, existing legal rights, or future conduct; and (b) in the circumstances the inducing party ought reasonably to expect that the relying party might act in reliance on the assumption in such a way that he or she will suffer detriment if the inducing party behaves inconsistently with that assumption; and (c) the relying party does act on the assumption in such a way, then it is unconscionable for the inducing party to act inconsistently with the assumption, at least without taking steps to ensure that the relying party suffers no detriment as a result of the action he or she took in reliance on the assumption.Footnote 106
Application of the doctrine of estoppel in the case of NOMs is an occasional application. Yet it raises a concern that use of the doctrine will eventually drain the NOM clause of content, even in situations in which, according to our analysis, it is necessary to grant legal validity to the NOM clause. In MWB v Rock, the Supreme Court was aware of this concern, and therefore, it cautioned that such use should be moderate.Footnote 107 At the same time, the Supreme Court did not clarify the criteria that would allow the use of the doctrine of estoppel as a means of reducing the injustice that enforcement of the NOM clause can produce, on the one hand, and not render the clause void, on the other. In this section, we seek to fill the gap with two criteria that limit the breadth of cases in which the doctrine of estoppel can be used.
First, the doctrine of estoppel should be applied as a rule only to the past. To the extent that the case involves a party that has deviated from the written contract based on representations, promises and actions of the other party, it is justified to apply the doctrine of estoppel and to prevent the other party from suing for the past deviation despite the fact that the contract contained a NOM clause.Footnote 108 In contrast, a demand of one of the parties, based on the doctrine of estoppel, to continue deviating from the contract in the future, or alternatively, a demand from the other party to continue to provide services that were not required by the written contract, should be rejected. In this way, a difference is created between cases in which the NOM clause is not enforced, and therefore the conduct that deviates from the written contract is granted full contractual validity, and cases in which the NOM clause is enforced, and therefore the conduct that deviates from the formal contract is granted only limited validity by virtue of the doctrine of estoppel.Footnote 109
Second, while in English law, estoppel itself can be used only as a defence, in other common law systems, it can be used to found a cause of action.Footnote 110 In such legal regimes, although generally the doctrine of estoppel should not be applied regarding the future, it may be called for in situations in which reliance on false promise regarding the future causes damage. Nevertheless, according to our proposed regime, in this context, the result of the estoppel doctrine should be limited.Footnote 111 While generally the remedies for damage due to breach of contract include anticipatory damages, when the claim is based on the doctrine of estoppel, we propose that only reliance damages should be recognised.Footnote 112 In other words, the injured party should be entitled to compensation for damages related to investments or expenses that have resulted from the presentation of future conduct by the other party, but the doctrine of estoppel should not allow one party to realise expected profits due to modification by conduct that contradicts the NOM clause. Therefore, the doctrine of estoppel should be applied in this case in narrow form. To conclude, in some circumstances our proposed approach opens the door to the application of estoppel as a claim that creates a new right even in the future, but restricts it to reliance compensation in the event of a breach. Our approach differs, therefore, from the prevailing opinion in US case law that tend to award expectation damages, rather than restricting awards to reliance damages.Footnote 113
IV. Conclusion
The conventional debate on NOM clauses focuses on the question of autonomy, namely the right of parties to limit themselves in the future, or alternatively, to deviate from their previous agreements. The presentation of this debate often creates a sense of dichotomy that lacks sensitivity to nuance, circumstance and context.
At a practical level, on the one hand, doctrines such as estoppel have undermined the full enforcement of NOM clauses. On the other, the need to present evidence of the parties’ desire for permanent and enforceable change has made it difficult to grant validity to modifications by conduct, even when the alleged NOM clause was apparently not enforced. The combination of these limitations to the apparently sweeping positions have blurred the practical difference between the positions, in a way that has created a dissonance between the theoretical poles, and a frequent practical overlap.
This article has sought to introduce a change to both the theoretical and practical levels. On a theoretical level, the paper shows how the controversy over NOM clauses reflects a profound disagreement between relational contract theory and neo-formalist theory about the balance between the written contract and contractual relations.
Based on a novel theoretical construction, the article presents a contextual and nuanced model.
The model we propose distinguishes between three relationships between: (1) sophisticated and evenly matched parties; (2) unsophisticated and evenly matched parties; and (3) parties of unequal power, taking into account the identity of the party for whose benefit the NOM clause works. For each one of these relationships, the model answers three questions:
(1) Should the NOM clause be enforced?
(2) Is the doctrine of estoppel to be applied, and if yes, to what extent?
(3) How are auxiliary tests to be implemented to determine whether conduct indicates a desire for legal change?
Figure 1 represents the algorithm created. We believe that if it is adopted by lawmakers, it will contribute to more adequate regulation of the domain.
Figure 1. Proposed Model for Treatment of NOM Clauses
The practical model proposed in the paper is based therefore on two elements: (1) a distinction between different types of relationships; and (2) auxiliary tests which, in light of the circumstances of the case, examine whether the parties’ behaviour reflects a desire for legal change. Although we have sought to offer clear categories, and auxiliary tests that are conveniently applied, we do not deny that in some cases the decision as to which category of relationship a particular case belongs is not clear cut, and there are grey areas. Certainly, the application of the auxiliary tests regarding the intention of the parties requires judicial discretion, and as such, is not accurately predictable.
Compared to existing models, the model we propose may often lead to better results that are more sensitive to nuance. At the same time, there is no denying that it entails costs related to implementation, and perhaps adversely affects legal certainty. At first glance therefore, the proposed model, which is nuanced and difficult to apply, even if it may lead to more just results, may seem inferior to a simple dichotomous decision between enforcement or non-enforcement of NOM clauses. However, a deeper look shows that it is inaccurate to present the choice facing the designers of law as a choice between a sensitive and nuanced but difficult to implement model, and “rough-and-ready” decisions that are easy to implement. For example, as noted at the beginning of this chapter, the neo-formalist approach that required recognising NOM clauses in every case opened the door to non-recognition by agency of the rule of estoppel. However, the question of when the rules of estoppel are applied, and how the general policy in favour of enforcement is maintained notwithstanding those rules remains unclear. Thus, a presentation of certainty is created that does not reflect the fact that actual implementation of the decision remains in the hands of the judge without an unequivocal guiding criterion. At the same time, the relational contract approach that supports non-enforcement of NOM clauses, and the recognition of contractual modification by conduct, still forces the courts to decide if the parties actually intended to introduce a permanent legal change by their conduct. The approach does not present a criterion according to which the judge can decide what the parties intended in any given case. Therefore, existing approaches make it difficult not only to achieve just and case-tailored results, but also to achieve efficiency and certainty.
Thus the approach proposed in this paper, which acknowledges the complexity of cases, and establishes distinctions and auxiliary tests, may in some contexts not only add nuance and solutions tailored to the situation, but also present more certainty than existing alternatives.
