Introduction

The doctrine of consideration has long been a source of criticism and antipathy, especially when contract variation is at stake. While consideration in general has remained, in Lord Denning’s phrasing, “too firmly fixed to be overcome by a side-wind,”[1] years of persistent ill wind have nevertheless eroded the consideration requirement for contract variations. The consideration requirement, or the pre-existing duty rule, has long been criticized as a cumbersome tool for preventing commercial extortion or duress. The pre-existing duty rule has been characterized as both overinclusive and underinclusive.[2] Overinclusive because the rule may endorse promises brought about through coercion, but which also happen to have some consideration present. Underinclusive because it precludes gratuitous promises that were sensibly and freely given. A growing impression shared by many is that the pre-existing duty rule is inconsistent with modern commercial reality and the legitimate expectations of business parties.[3] Given the persistent dissatisfaction with the rule, it is perhaps unsurprising that the doctrine of economic duress has steadily come to displace the consideration requirement in many common law jurisdictions.

The leading case of Williams v. Roffey[4] initiated the shift away from the pre-existing duty rule associated with Stilk v. Myrick.[5] Appellate courts in New Zealand and Canada have since gone further, dispensing with the consideration requirement altogether to fully embrace the doctrine of economic duress.[6] The view that a promise to vary a contract should be enforceable so long as it was not procured under economic duress will accordingly be labelled the economic duress model. While the economic duress model provides advantages over the classical approach of Stilk v. Myrick, there are substantial shortcomings that should cause a reconsideration of the trend toward its embrace.

An initial difficulty is that economic duress has not been definitively expressed as a replacement for consideration, but rather as merely an alternative means of determining enforceability. The ungainly patchwork of rules governing contract variation has thus only been expanded. Various doctrines currently sit alongside one another, with their respective use often contingent upon distinctions between promises for more or promises for less performance.[7] In Canada, a purported contract variation could potentially be analyzed simultaneously through four doctrines or rule sets, including: common law consideration under either Stilk v. Myrick or Foakes v. Beer, the doctrine of economic duress, promissory estoppel under equity, or part performance permissible under statute.[8] Such an overlapping mix, containing both redundancies and contradictions, can hardly be said to be conducive to contractual certainty or economic efficiency.

A model based on economic duress offers the prospect of promoting socially useful variations that reflect the reasonable expectations of the parties; however, this must be balanced against the potential for incentivizing opportunistic renegotiations, and undermining the pricing function of contracts in the first instance. Another challenge is that economic duress is not nearly as straightforward or clear-cut as it may at first appear. Determining what constitutes legitimate or illegitimate pressure in a commercial context can be quite subjective and open to interpretation. The most significant concern with the economic duress model addressed here, however, is the potential for it to undermine the complementary role of promissory estoppel in protecting parties from unfair retractions of promises. The economic duress model may tend to be overly inclusive in finding promises enforceable. In particular, it risks transforming gratuitous indulgences meant only to temporarily suspend contract terms into binding contract variations. In other words, the economic duress model may transform a commercial favour or leniency into an entitlement. The move toward the economic duress model threatens the distinction between promises that are deserving of an equitable protection and promises that are deserving of becoming an irrevocable contractual right or entitlement.

It is proposed that a model based on the changed expectations of the parties may serve to maintain the distinction between promises that vary or alter contractual rights and those promises that merely suspend them, such as relaxations or indulgences, that may be binding but which do not change contractual terms. It is also suggested that a model of changed expectations can provide an improvement upon both the pre-existing duty rule and the model of economic duress, while also covering the role played by promissory estoppel in a more coherent and efficient fashion.

Under a model of changed expectation, the promisor changes the reasonable expectations of the parties with a promise, whether gratuitous or backed by consideration. What determines how permanent the change may be, or whether it is a relaxation or a variation of contractual terms, becomes a question of how the promise was initiated or provided for. The key distinction becomes not enforceability of a post-formation promise, but the availability of revocation. The proposed model of changed expectations would hold that all seriously intended promises that alter the expectations of the parties are enforceable, subject to revocation. Revocation of post-formation promises must occur in the same manner as the promised variation. Bilateral promises, in which both parties alter their promised performance, as with consideration, will require both parties to consent to the revocation and the return to the original terms. Unilateral promises, which are given gratuitously by one party alone, may be revoked unilaterally by the promisor subject to reasonable notice. The first level of promises, termed bilateral, is essentially a traditional inquiry into the presence of consideration. The modest, or incremental, innovation proposed here occurs on the second level of promise, labelled unilateral, wherein gratuitous promises become binding but are subject to unilateral revocation.

The proposed approach draws on the flexibility of promissory estoppel, in that the original terms can be reinstated. The difference under the proposed model of changed expectations is that only the issue of notice would be assessed, not whether the return would be inequitable or not. This streamlines the inquiry into enforceability based on timing, dispensing with the need for a qualitative assessment of fairness between the parties, or an assessment of whether the promisee relied on the promise to their detriment. Also jettisoned would be the differing treatment between promises for more and less, and the traditional restriction that promissory estoppel can only serve as a shield and not a sword.

The proposed model of changed expectations would: i) streamline the role previously played by equity; while ii) being far more generous in enforcing seriously intended promises than the classical approach of Stilk; but iii) remaining less generous than the economic duress model which may freely transform all serious promises into contractual rights. Unlike the economic duress model, under a model of changed expectations, not all seriously intended promises are treated equal. Gratuitous promises depend for their enforceability on having performance occur before the promisor revokes the promise with reasonable notice. The gratuitous promisor who decides to pursue the original terms after the promisee has tendered performance will be out of luck. For example, the gratuitous promisor who revokes (with reasonable notice) halfway through delivery instalments will have to pay for half of the deliveries based on the changed expectation rate and half based on the original rate, and so on.

The proposed model of changed expectations would distill the inquiry into enforceability into two questions facing a promisor who wishes to resile from a post-formation promise: 1) Was the promise gratuitous? 2) Is there a reasonable time to change expectations back to the original terms before performance? If both questions are answered in the affirmative, then the promisor may revert to the original contract’s terms for the performance that remains after a reasonable revocation.

Part I of this article examines the historical connection between the pre-existing duty rule and the policy objective of protecting parties from undue commercial pressure or duress after a contract has been formed. Part II addresses the modern move away from the consideration requirement in favour of the doctrine of economic duress. Part III examines contract variation from an economic perspective. Part IV elaborates on the disadvantages of an overly permissive approach to contract variation, with particular emphasis on the Rosas v. Toca decision of the British Columbia Court of Appeal. Part V explains a model of changed expectations, and attempts to demonstrate its advantages over both a model of economic duress and that of the previous status quo based on consideration and promissory estoppel.

I.   Consideration, Duress, and Discontent

Consideration provides a basic evaluative function for determining the enforceability of promises under the common law. Ideally, consideration distinguishes purely gratuitous promises from those which resemble a bargain, or tie a provided promise to some contingency, or direction of value, involving the promisee. A well-known definition of consideration was given in Currie v. Misa:

A valuable consideration, in the sense of the law, may consist either in some right, interest, profit, or benefit accruing to the one party, or some forbearance, detriment, loss, or responsibility, given, suffered, or undertaken by the other.[9]

A bargain model of consideration was reflected in the treatment of contract variation as well, in the so-called pre-existing duty rule, or the rule in Stilk v. Myrick.[10] According to the pre-existing duty rule, one’s satisfaction of the consideration requirement cannot be met by performing what is already owed under the contract. In Stilk, after two crew members had deserted a ship mid-voyage, and while in a subsequent port, the ship’s captain offered to pay the remaining crew members increased wages if they stayed on and sailed safely back to London. The ship was not in immediate peril at the time of the promise, nor was the promise extracted by any threat from the crew members. Nonetheless, the promise of increased wages was found to be unenforceable. The increased work required to cover for deserted crew members was determined to be part of the usual exigencies of such precarious voyages, and therefore the performance of what was already contracted for.[11]

The pre-existing duty rule has long been interpreted as a proxy for the policy goal of preventing commercial extortion. Many early and notable cases featuring the pre-existing duty rule emerged in a maritime context, in which the “hold up” prospects are vivid.[12] As Lord Kenyon cautioned in Harris v. Watson,[13] to enforce promises of ship captains to pay sailors extra wages for doing more than their ordinary share of duties would open the prospect of sailors threatening to allow the ship to founder or sink lest the captain accede to their demands, no matter how extravagant.[14] Notably, the circumstances in Stilk v. Myrick did not mirror the flagrant threat contemplated by Lord Kenyon, as the captain proposed the increase when the ship was safely in harbour.[15] Nor did the reasons of Lord Ellenborough proceed on the same express policy concerns for Britain’s maritime economy and defences: “I think Harris v. Watson was rightly decided; but I doubt whether the ground of public policy, upon which Lord Kenyon is stated to have proceeded, be the true principle on which the decision is to be supported. Here, I say, the agreement is void for want of consideration.”[16]

The direct connection between consideration and the policy goal of preventing duress is intuitive and persistent. If one might quibble with treating consideration as exclusively synonymous with preventing commercial extortion, this is nevertheless a valuable policy goal. There is a prospect that individuals will opportunistically take advantage of the other contract party’s post-formation financial vulnerability to extract additional proceeds. Such proceeds are not due to any change in circumstances or additional contribution, but rather simply the ability to extract more on a “hold up” basis than was available ex ante when the promisor was at liberty to pursue other options. As Posner has observed: “[T]he making of a contract may confer on the seller a monopoly vis-à-vis the buyer which the seller can exploit by threatening to terminate the contract unless the buyer agrees to pay a higher price than originally agreed upon.”[17]

The rule in Stilk dictates that in order to receive a new benefit under a pre-existing contract, one must provide new or fresh consideration in exchange. This undeniably produces a certain rigidity or inflexibility in that each contract variation must be held to the same evaluative or qualitative standard as the formation of the initial contract. The pre-existing duty rule has produced great consternation amongst those who hold that individual autonomy, or the individual will, requires giving effect to seriously intended promises;[18] as well as those who desire a more tort-like or neighbourly expression of contractual obligations;[19] and those who indicate that the gift filtering function is unnecessary when there is an ongoing commercial relationship.[20]

Despite the many objections that the rule in Stilk has garnered, it may be said that there is nonetheless a moral quality in the consistency of the consideration requirement. Whether one agrees with the premise of the rule in Stilk or not, consideration does provide a clear answer to why gratuitous promises are not to be enforced, whether between strangers or contract parties—because nothing was given in exchange for the benefit. A donative or charitable pledge may be seriously intended, and yet many might think that it should not give rise to a positive obligation to deliver unless the promisee has done something in exchange for the commitment. Regrettably, this notion of what is deserved or earned by the promisee has been shuffled out of view with the move away from the consideration requirement. In the leading cases which adopt the doctrine of economic duress as the central enforceability standard for contract variations, the move is decidedly undertheorized. Why is the doctrine of economic duress preferable? Because the rule in Stilk v. Myrick is antiquated and the new rule allows for business people to arrange their own affairs? This sounds sensible enough, but on closer scrutiny the question of whether a benefit is deserved may still have resonance.

II. Practical Benefit and Economic Duress

One of the most notable judicial[21] steps away from the traditional pre-existing duty rule occurred in the English Court of Appeal’s decision in Williams v. Roffey. A general contractor had undertaken to renovate a block of flats, for which they contracted with various subcontractors to do portions of the work, including the plaintiff carpenter. The defendant general contractor worried that delays in the carpentry work would delay the entire project, and render them liable to a financial penalty on the wider project. To avoid this prospect of delay, the defendant agreed to pay the carpenter more money to complete the carpentry work. At this point this would appear a classic situation of a gratuitous promise that would be unenforceable according the rule in Stilk: there was no change in circumstances other than the poor performance of one of the parties, and a wider project that may be held up to ransom makes the promisor vulnerable in a way that they were not a formation. Some notable features, however, provided an avenue for the court to step in a new direction: the original price agreed on for the carpentry work was known by the defendant to be unrealistically low; the carpenter was bumbling and inefficient, but apparently bumbled in good faith and made no threat or demand for more money; and, most importantly, the more powerful and cognizant party voluntarily approached the carpenter and offered to pay more.

In Williams v. Roffey the court famously concluded that a practical benefit to the promisor could constitute good consideration. Or, put differently, a promised variation would not necessarily fail due to want of fresh consideration provided or promised by the promisee in return. One may speak of mutual benefit, as did Lord Purchas, but this is redundant since it goes without saying that the plaintiff suing for more stands to gain a benefit from the defendant’s promise to pay more. The radical departure from Stilk is that the practical benefit does not have to flow from the plaintiff or promisee, and indeed the benefit to the defendant promisor may simply be the avoidance of the plaintiff’s possible breach. Notably, it would not have to be an actual breach. Avoiding a single uncomfortable phone conversation with the other party could conceivably constitute practical benefit to the promisor and thus render a gratuitous promise enforceable.

Interestingly, all three sets of reasons in Williams v. Roffey strain to assure that they are not overturning the rule in Stilk, while also characterizing Stilk as outmoded and confined to its particular context of involving “the rigours of seafaring life during the Napoleonic wars.”[22] Perhaps the most common tactic involved in the characterization of Stilk is to treat it as equivalent to the policy goal of preventing duress. As Lord Glidewell wrote:

an agreement to pay an increased price may well be voidable because it was entered into under duress. Thus this concept may provide another answer in law to the question of policy has troubled the courts since before Stilk v. Myrick, and no doubt led at the date of that decision to a rigid adherence to the doctrine of consideration.[23]

While policy concerns may certainly have driven the result in Stilk, again it is conceivable to recognize a wider view to moral desert as well.

If Stilk is simply a stand-in for preventing duress, then modernizing the common law so it becomes a straightforward inquiry into the presence of duress is tempting. Drawing on Lord Scarman’s utilization of economic duress in Pao On,[24] economic duress was used by the court in Williams v. Roffey as the ultimate determinant of enforceability.[25] If, inter alia, the promisor obtains a practical benefit—including from someone other than the promisee—then so long as the promise was not procured by economic duress or fraud on the part of the promisee, the variation will stand. The practical benefit test for consideration enunciated in Williams v. Roffey marked a notable shift in focus from the promisee to the promisor. This new focus would subsequently come to define the model of economic duress. For nearly two centuries, the common law had focused the evaluative inquiry squarely on what the promisee had offered in exchange for a promise. Furthermore, the expansion of promissory estoppel had similarly focused squarely on the promisee or recipient of a gratuitous promise; namely, on why it would be unfair to return to the original contract’s terms. The established common law inquiry into whether the promisee had promised anything in return for a benefit was discarded in favour of a presumptive entitlement for the promisee, so long as the promisee had not used coercion to extract the benefit.

While the practical benefit test remains alive in the United Kingdom,[26] and has been followed in Australia,[27] courts in Canada and New Zealand have seemingly leapfrogged the step altogether in order to do away with the consideration requirement for contract variation. Bypassing the practical benefit is understandable given that the test is in effect superfluous or redundant. From the outset the practical benefit test was met with skepticism. Indeed, the seeds of the problem are indicated in the decision itself. Part of the practical benefit ascribed to the defendant promisor was the avoidance of having to find a replacement contractor. Essentially, avoiding the prospect of the other party’s breach can count as consideration.

The problem of course is that it is difficult to conceive of a situation which does not provide some modicum of practical benefit.[28] For instance, any saved cost or inconvenience in even raising a complaint about delayed performance could surely count as a “peppercorn” of practical benefit. Unsurprisingly, the Williams v. Roffey decision has been controversial. Critics were dubious of the court’s assurances on the compatibility of the result with the traditional rule in Stilk v. Myrick. Critics instead recognized that the decision threatened to do away with the rule completely.[29] The criticism of the decision was not limited to academics, as lower courts have also expressed reservations or discomfort with the approach in Williams v. Roffey.[30]

If practical benefit may be found in every situation in which a contract is maintained and a breach avoided, then the approach in Williams v. Roffey inevitably funnels to the question of whether economic duress was present in extracting the promise. And if practical benefit is superfluous or redundant in this manner, then why not simply exchange the test for the presence of consideration for that of economic duress? This invitation has been accepted by the courts of New Zealand, and with growing momentum by Canadian courts. This article attempts to supply the answer as to why courts should not simply embrace economic duress; but, first, it will summarize the evolving positions in the case law.

The first and most notable New Zealand case moving beyond the practical benefit step was issued by the Court of Appeal in Antons Trawling Co Ltd v. Smith.[31] Justice Baragwanath, writing for the court, held that intentionality, not consideration, was decisive in determining whether a promised contractual variation was binding or not. Justice Baragwanath signaled that the promisee’s reliance was required in addition to the promisor’s agreement: “[W]here parties … have acted upon an agreement to a variation, in the absence of policy reasons to the contrary they should be bound by their agreement.”[32] While the “acted upon” requirement seems to echo the reliance protection of equity, the case has been treated as supporting the notion that consideration is no longer definitive, regardless of practical benefit. As Justice Baragwanath himself clarified: “[C]onsideration is not required in the case of an existing contract where there is no evidence of oppression.”[33] As Marcus Roberts has observed, a number of members of New Zealand’s High Court have cited with approval Justice Baragwanath’s position in Antons Trawling. Until New Zealand’s Supreme Court provides a definitive answer, it seems that “the lower New Zealand courts will not require evidence of consideration, whether of the traditional or the practical benefit variety” in determining the validity of a contract variation.[34]

The position in Canada is similar to that found in New Zealand. While there has been no central decision clarifying the state of the law from Canada’s Supreme Court, an appellate decision that dispenses with the consideration requirement in the “absence of oppression” has had a growing influence. In Greater Fredericton Airport Authority v. Nav Canada,[35] the New Brunswick Court of Appeal was faced with a situation in which a contractor threatened that, unless the owner paid for the cost of an equipment replacement, the contracted-for runway construction would be delayed. The airport authority disagreed strenuously, and only signed the agreement under protest. While the agreement would easily have failed for want of fresh consideration, it also bore some of the hallmarks of an agreement entered into under duress. Justice Robertson, writing for the court, took the opportunity to declare that “there are valid policy reasons for refining the consideration doctrine to the extent that the law will recognize that a variation to an existing contract, unsupported by consideration, is enforceable if not procured under economic duress.”[36] Chief among the “policy” reasons enumerated by Justice Robertson were commercial efficacy and the legitimate expectations of the parties in rearranging their deals, as well as the need to move beyond the antiquated context of Stilk v. Myrick. And echoing the judicial diplomacy encountered in Williams v. Roffey, Justice Robertson also stressed that his “incremental approach” was by no means an “abrogation of the rule in Stilk v. Myrick.”[37] Yet, if the result in Williams v. Roffey was open to the charge of being far more radical of an undermining of Stilk v. Myrick than adverted to, then the same argument holds even more strongly for Nav Canada.

Such judicial diplomacy may no doubt be inevitable whenever common law innovations percolate upwards, rather than in top-down hierarchical pronouncements from an apex court or through legislative reform; however, there is a certain ambiguity promoted by this diplomacy. The suggestion that consideration persists alongside economic duress only invites lower courts to waffle in the ambiguous wake of what are otherwise bold cases. The ambiguous position of Nav Canada was only increased by the fact that only a year later the very same appellate court, with the very same Justice Robertson writing for the court, distinguished Nav Canada on the facts and inquired into the presence of traditional consideration in order to determine whether a contract variation was binding.[38]

The initial judicial reception of Nav Canada appeared uncertain at best, including tepid influence amongst lower courts in Canada, and the Alberta Court of Appeal rejecting the approach in Nav Canada outright.[39] It thus appeared that the decision might be left as an outlier to be eventually reabsorbed back into the common law mainstream—until the decision of the British Columbia Court of Appeal in Rosas v. Toca.[40] In Rosas, the court declared that: “When parties to a contract agree to vary its terms, the variation should be enforceable without fresh consideration, absent duress, unconscionability, or other public policy concerns.”[41] The public policy reference is not explicated at all, but it is in keeping with a similar reference by Justice Baragwarath in Antons Trawling, which went similarly unexplained.[42]

Justice Bauman, writing for the court in Rosas, observed the “cases that have adopted reforms to the doctrine of consideration appear to focus on the seriousness of the parties’ intentions and the legitimate expectations of business parties.”[43] Then, following Nav Canada in both substance and style, Justice Bauman invoked language of an “incremental change” in the law, and the need to fulfill the “legitimate expectations” of the parties.[44] And again, assurances were made that the decision should not be read as overturning the rule in Stilk. Indeed, apparently a “variation supported by valid consideration may continue to be enforceable for that reason, but a lack of fresh consideration will no longer be determinative.”[45] Yet, why bother to utilize a rule if it is no longer determinative? From Williams v. Roffey through to Rosas v. Toca, few have found the judicial language of incrementalism convincing. As Marcus Roberts observed when reviewing Rosas v. Toca: “The courts are in reality enforcing all variation agreements as long as they are freely agreed to.”[46] The question becomes whether this approach of enforcing all freely agreed to variations is actually an improvement over the consideration requirement that it replaces.

III.  Contract Variation in an Economic Perspective

There are numerous theories or frameworks for evaluating and scrutinizing the enforceability of contracts, but an economic perspective is arguably unique in providing a view to not only how legal rules create incentives between individual parties, but also a view to whether these rules and incentives are productive for society in the aggregate. Of particular interest for this investigation is how rules for contract variation permit unforeseen post-formation developments, and whether these rules are conducive to an efficient allocation of risk between the parties.

As with many areas of law subjected to an economic analysis, the most influential contribution on contract variation remains that of Posner. Posner’s framing of whether variations ought to be enforceable contains two main categories of cases: those in which the basic underlying circumstances are unchanged since formation; and those in which there has been some material change outside of the control of the parties, or in the surrounding economic environment.[47] Posner’s framing could be said to reflect the modernization of the law of contract variation, moving from priority one in the prevention of duress toward priority two and the appeal of socially useful adaptations to changed circumstances.

When there are no changed circumstances external to the parties that justify adapting the terms of the deal, there is seemingly no rationale for the varied bargain other than duress. For why would any rational self-interested person agree to pay more or accept less unless in exchange for something, or a sufficient quid pro quo?[48] This recalls Posner’s characterization of extorted variations as a form of abuse of monopoly power, in which the extracted surplus is of no social value.[49] Posner uses the case of Alaska Packers’ Ass’n v. Domenico[50] to demonstrate an example of a monopolistic variation that was properly deemed unenforceable by the courts. The plaintiffs, or promisee(s), had agreed to serve as sailors and fishermen but had contracted in writing for a stated term and compensation level. But soon after the contract began, and at a point when it would have been impossible for the defendant to secure replacements, they refused to continue to perform unless they were paid more.[51] When there is no material change outside of the parties’ control and one party only acts to opportunistically seek more gain because it suits them to do so, it is a monopolistic exertion of pressure that yields no added social value. Such promises should be unenforceable on economic grounds and, in categorically declaring that such promises were unenforceable, the rule in Stilk was efficient.

When circumstances have changed, however, there is the potential for socially useful bargains and contract variations unsupported by consideration. The notion of changed circumstances must go beyond the performance or plans of the contract parties themselves, however, for simply deciding that one wants more money would render the distinction meaningless. Similarly, the changed circumstances must include only new developments beyond the reasonable control or planning of the parties. If a buyer makes independent plans and investments based on the expectation of receiving an item promised for delivery by the seller, these are changed circumstances that would explain the buyer’s vulnerability but would not be justified since they are not unanticipated or uncertain post-formation developments. When changed circumstances are external to the parties and beyond their control, as in changes to the overall marketplace, it presents the potential that both parties can realize a benefit and that it would be socially valuable to have the variation stand even if it is not supported by consideration. In Williams v. Roffey the change of circumstance was clearly of the internal variety: the avoidance of the penalty clause to the third-party owner, which provided a benefit to the promisor, would surely have been present at formation; the new challenge that emerged post-formation was the poor planning and underperformance of the other party, the subcontractor.

In abstraction, or in a situation of little to no transaction costs, one could imagine a circumstance in which both parties could costlessly return to the ex ante position and would, in light of the new information available, choose to contract at a higher price with the same contract party again. Which is to say that if an auction, for example, could be held a second time after an unknown variable is revealed, the same bidder would offer the highest bid both ex ante and ex post. This is a practical benefit in a fuller sense than the avoidance of breach suggested in Williams v. Roffey, in that it is as if both parties find it desirable to contract again, as if de novo, with the same contract party. The key is that the external environment has changed, not merely the calculations of the parties.

Within changed circumstances, one can discern two subsets of example types: i) newly discovered circumstances, and ii) a substantially altered marketplace or economic environment.

A common example of newly discovered circumstances is that of a contract for soil excavation, which turns out to be a much more onerous undertaking once digging begins. By contrast, the latter category of changed circumstances will tend to focus on factors beyond the immediate control of the parties, such as unpredictable scarcity which makes delivery very difficult or next to impossible. A famous environmental change of circumstance case in which the promise was held to be enforceable is the American case, Goebel v. Linn.[52] The defendants were brewers who contracted with the plaintiffs for the supply of ice that was to be used to refrigerate the defendant’s beer. When a milder than usual winter meant that the plaintiff’s ice harvest was far less than anticipated, the plaintiff informed the defendant brewers that the agreed-upon ice could not be delivered at the agreed-upon price. The defendants had a large stock of beer that was at risk of spoiling, and they therefore agreed to pay nearly double the contract price to ensure the promised delivery of ice. The defendants later repudiated the agreement and the court found for the plaintiff, observing that the defendants:

Chose for reasons which they must have deemed sufficient at the time to submit to the company’s demand and pay the increased price rather than rely upon their strict rights under the existing contract. … Unexpected and extraordinary circumstances had rendered the contract worthless; and they must either make a new arrangement, or, in insisting on holding the ice company to the existing contract, they would ruin the ice company and thereby at the same time ruin themselves.[53]

While Goebel has been criticized for condoning commercial extortion, Posner nevertheless endorses the decision as a justifiable variation on the grounds of changed circumstances.[54] Posner emphasizes that the defendant in Goebel, unlike the opportunistic advantage seeker in Alaska Packers’, was involuntarily withholding performance due to a change in circumstances beyond their control.[55] While one might have greater sympathy with a party that faces an external change of circumstances that is beyond its control, such classifications are inherently interpretative and subjective. Ultimately, Posner’s framing is problematic as a result. Furthermore, the malleability in determining what constitutes duress or what counts as legitimate commercial pressure is reminiscent of the problems in interpreting consideration. Indeed, it has been observed that the jurisdictional experience in applying economic duress has proven to be given to interpretive uncertainty reminiscent of the criticisms of consideration.[56]

The underlying question in such cases is seemingly one of risk allocation. Namely, was the new circumstance part and parcel of the supplier’s overall risk, or was it something truly unforeseen and therefore excusable as a reason for non-performance? The shortage of ice may be a once-in-a-lifetime event that leaves the unfortunate supplier in a precarious position, or it may be a regular occurrence that should be priced into the regular agreement to supply ice. The parallels with the doctrine of frustration are apparent here, as others have observed.[57] There are times when there are completely unforeseen events, or force majeure situations, which are deemed to absolve parties of any further performance or liability.[58] Yet not every uncertain development is completely unforeseen, and courts are generally minded to determine whether one party implicitly undertook greater responsibility for the uncertainty and its attendant risks.[59] This view to risk allocation is significant in recognizing that the division between changed and unchanged circumstances, between monopolistic duress and legitimate variation, is superficial and incomplete. An overall view to efficiency or social utility must take risk distribution into account, both ex ante and ex post. Just as it would be ill advised to treat all unforeseen events alike in absolving future contract performance, so too should changed circumstances be viewed skeptically.

A useful layer of nuance was added to the economic analysis of contract variation in an article by Aivazian, Trebilcock, and Penny, wherein the authors distinguish between static and dynamic efficiency.[60] Static efficiency includes the immediate practical concerns of the contract parties, whereas dynamic efficiency involves the long-term incentives that are informed by legal rules. Significantly, the shift of risk bearing that can occur with a contract variation, even when it is procured without duress, can alter the overall efficiency equation. If a contract party is able to readjust the assignment of risk between the parties so that the risk of a mild winter and less ice is borne by the buyer and not the seller as originally implied in the contract, this introduces the problem of moral hazard.[61] If the risk of unforeseen or unknown circumstances can be shifted through contract variation, this lessens the incentives for parties to contract prudently and accurately determine their best firm price in the first instance. A permissive view to variation allows for an escape from lax or overly generous offers and disincentivizes firmly established risk allocation. It may be tempting to enforce all seriously intended promises under the guise of the “legitimate expectations” of the parties, but the reality of costless risk spreading that occurs between ex ante and ex post contract formation should suggest that gratuitous variations should be viewed skeptically, on both economic and moral grounds.[62] Risk allocation between the parties is addressed further below as an important feature of the model of changed expectations.

Any benefits of contract variation that account for changed circumstances may well be overcome by the greater social detriment in the cost of contract uncertainty, as one contract party will be able to costlessly increase risk and its cost to the other contract party. Indeed, the significance of this potential for risk redistribution through contract variation is such that Aivazian, Trebilcock, and Penny conclude:

Only in cases where the efficient allocation of risks is indeterminate, both subjectively and objectively, or where the risk in question is extremely remote so that the expected costs of bearing it do not induce significant efficient precautionary responses, is it likely that the static efficiency gains from recontracting will outweigh the dynamic efficiency losses from permitting the reallocation of risks through variations exacted and acceded to.[63]

The simple binary presented by Posner of duress and socially desirable contract variations is drastically circumscribed when factoring in risk distribution. The areas of potential utility of gains are seemingly limited to cases in which the changed circumstances are so remote that the doctrine of frustration would excuse future performance because neither contract party implicitly assumed greater responsibility for such a rare occurrence. This severe restriction on the efficiency gains of a permissive model of contract variation is only exacerbated when more informal modes of contract variation are considered. The following section elaborates on the disadvantages of an overly permissive approach to contract variation, with a particular emphasis on the decision of the British Columbia Court of Appeal in Rosas v. Toca.[64]

IV. The Moral and Economic Arguments for Contract Relaxation

      Rosas v. Toca is not only notable for helping turn the tide of Canadian jurisprudence in the wake of Nav Canada, it is also indicative of the serious and overlooked flaws that lie at the heart of the economic duress model. The contract in Rosas v. Toca emanated out of friendship and was highly informal. The plaintiff had won a multi-million dollar lottery, and among various financial gifts advanced to her friends, the plaintiff also agreed to lend the defendant $600,000 toward the purchase of a new home.[65] No part of the agreement was recorded in writing and the trial judge had to find not only the existence of the contract itself, but also that the loan was implied to be interest free (presumably because it was never mentioned by the parties themselves), and that the time limit for the loan was put at one year.[66] The plaintiff requested repayment multiple times but at each instance the defendant requested more time, to which the plaintiff acquiesced quite readily as she had no pressing need for the money.[67] The difficulty for the plaintiff was that by the time that the friendship had soured and the plaintiff brought suit for the outstanding amount, the statutory limitation period on suits for the repayment of a debt had run out.[68] Unless there was a contractual variation at common law, as opposed to an equitable relaxation, the plaintiff would be barred from recovery and the defendant would receive an unjust windfall, transforming an interest-free loan into an outright gift. The difficulty of course is that under the traditional rule in Stilk, the contract variation would require consideration, and there was clearly none—the defendant had provided nothing in exchange for each extension.[69] It would be manifestly unfair if the defendant were able to rely upon her lack of consideration, or promised quid pro quo, to transform a generous loan into a substantial gift.

The reasons of the court in Rosas, delivered by Chief Justice Bauman, begin tellingly with a quote from Mr. Bumble in Dicken’s Oliver Twist, who laments that “if the law supposes that … the law is an ass.”[70] Things were not looking good for consideration, then. Interestingly, the court did not determine which of the laws at hand were responsible for the law looking like an ass, because it might well be said that it was the limitation period, rather than any dictate of the pre-existing duty rule, that would have yielded the unfair outcome. In any event, the court ultimately determined that the common law consideration requirement would have to give way.

The nature of the deal in Rosas v. Toca could be described as indulgence or unilateral grant of relief, especially as the loan emanated in friendship and not commerce, and would therefore normally fall within the province of equitable relief and promissory estoppel. Yet since the parties did not pursue a claim under estoppel[71] and the procedural necessity of the limitation period could only be satisfied by a more formal contract variation, only a common law contract variation could yield the desired fair outcome. What the case demonstrates, undoubtedly as an unintended consequence, is the potential for the modern use of economic duress to disintegrate the established role played by equity. Equity provided a flexible, protective shield for circumstances that fell short of the formal or bargained requirements of common law consideration. Cases like Nav Canada and Rosas, by contrast, raise the prospect that every small relaxation or forbearance may constitute an irrevocable rewriting of the contract. The danger, furthermore, is that an increased number of unilateral relaxations in the performance of a contract will be treated as bilateral variations of the terms of the contract, meaning that the relaxation cannot be undone without the consent of the beneficiary or promisee. The issue of contract relaxation, including forbearance and waiver, illustrates that the modern approach to contract variation premised on the doctrine of economic duress is unsatisfactory from both a moral and an economic standpoint.

The economic duress model does have some potential safeguards against limitless or indefinite contractual rights, but these are arguably insufficient. The first safeguard lies in the interpretation of the parties’ intentions, especially that of the promisor. A simple finding that a favour or leniency was not seriously intended to have contractual effect could serve to exclude some undesirable liability for trivial assurances. Yet a gratuitous indulgence in accepting late payment from a tenant, for example, without an express promise present, would likely need to be at once binding and limited in application. The indulgence would need to be binding if it is found that there was the implication that the tenant was meant to avoid responsibility for the interest, for example.

A more wide-ranging safeguard, though, would be found in the role of implied terms, for the reason that it would often be necessary to find some time limit to an indulgence or gratuitous promise. It is possible to have implied limitation periods that are of an uncertain duration to cover High Trees-like circumstances where one party has encountered financial difficulty post-formation. In this fashion, a gratuitous promise to accept less rent from a newly unemployed tenant could be read to be limited to the duration of the tenant’s unemployment, so that when the tenant resumes employment the original lease terms are returned to. The issue would be complicated additionally should the issue be of the tenant’s unexpected good fortune, perhaps through an inheritance or lottery win. No doubt a court would simply read the implied term to be financial difficulty instead of unemployment alone. Nonetheless, it is possible to imagine some strained interpretations connecting a promised relaxation of terms with future changes in the parties’ fortune or circumstance. This safeguard of implied terms would be a useful tool, even though it could be somewhat artificial in its application of judicial hindsight.

While the Rosas case is undoubtedly an oddity in the high degree of informality and its distinctly non-commercial nature, forbearing to insist on a repayment deadline is obviously a more general or commonplace occurrence. To demonstrate the difficulty with the approach of Nav Canada and Rosas, let us consider a relaxation involving a property rental. Given that one of the most famous promissory estoppel cases involved a relaxation of rental rates, it may be a fitting comparison. Suppose that a tenant runs into financial difficulty and cannot pay the agreed-upon rental rate. The landlord owner sympathizes with the tenant’s predicament and offers to accept less rent, a late payment, or both. The tenant did not coerce the landlord, who in this case is in a much more powerful position. The seriously intended promise, with duress absent, would simply be enforceable according to the economic duress model of Nav Canada. And of course the situation would be replete with practical benefit for the landlord. The tenant stays, and the landlord does not have to suffer the inconvenience of enforcing an eviction or finding another tenant. And recalling Posner’s economic classification, one could point to the tenant’s unemployment as an unexpected development akin to an ice shortage. Additionally, to highlight the ambiguity of such assessments, one could equally add that it was surely the tenant and not the landlord who implicitly bore the risk of the tenant’s unemployment. The pragmatism of the Williams v. Roffey innovation is apparent here thus far. It would no doubt be sensible to everyone involved to permit the relaxation and to have the tenant remain on lesser rent. Yet working through the hypothetical rental scenario can quickly illustrate why the economic duress model and its over-receptiveness to contract variation is problematic.

Assume first that the tenant relies on the landlord’s assurances and pays less, remaining in the apartment for months when he or she otherwise might have sought to move elsewhere. After these months the tenant finds a good job and is returned to prosperity. Should the landlord be able to return to the original contract rental rate? The answer indicated by High Trees[72] would clearly be yes, and for multiple reasons: namely, that it was implied that the rental relaxation would only last for the extenuating circumstances that brought about the tenant’s financial difficulty; and, more importantly, because after the end of the extenuating circumstances, it would no longer be inequitable for the landlord to insist on a return to the original contract. It is arguably this latter feature and not the implied time limit that better encapsulates the spirit of promissory estoppel. The feature of a return to original terms with notice is important because it indicates that relaxation can expire or be withdrawn, with time to conclude reliance, on the sole discretion of the promisor. How would the economic duress model treat the situation? Likely a court would imply the same time limit, and the result would arguably look the same so long as the extinguishing event occurs before the landlord has had a change of mind. What is objectionable about the economic duress approach, however, is that it must, seemingly as a matter of logic and principle, prevent a return to the original contract terms before the end of the extenuating circumstances regardless of notice.

To stress the difference between a contract variation and a mere relaxation: with a relaxation, the original contract terms persist but are merely suspended. This distinction is more fully rendered when one imagines the landlord seeking a return to the initial contract terms before the end of the extenuating circumstances or any implied time period. Suppose that the landlord agrees to accept less rent, at which point a contract variation forms under the economic duress model. Quickly thereafter, in the same conversation and after only fourteen seconds, the landlord recollects some prior commitment and withdraws or attempts to revoke his promised willingness to accept less rent. Should the landlord be able to insist on the original rent, or can the tenant rely on the promise allowing him to pay less until he gets back on his financial feet? If the economic duress model is to be applied faithfully and honestly as a matter of contract variation, in which variation means that terms change, then the landlord cannot resile from his promise and the tenant will have the advantage of the promise until the implied time limit, set at the time of the second promise. It is difficult to imagine some implicit way out of the conclusion that the seriously-intended promise of the landlord was not a contract variation as dictated by Nav Canada and Rosas.[73] And perhaps advocates of the economic duress model would unabashedly celebrate the tenant’s contractual right. The court in Rosas was given the opportunity to address this potential to fully embrace the implications of enforcing every freely given promise to vary, but shrank from it. Specifically, Chief Justice Bauman stated that it “was unnecessary to consider whether Ms. Rosas would have been prevented from bringing an action within the forbearance period.”[74] As a practical matter, yes, it was unnecessary in that Ms. Rosas obviously never attempted to withdraw the promised forbearance and hence ran afoul of the limitation period. As a matter of principle and generalizable lessons from an appellate review, however, this answer is quite unsatisfactory. The court’s determination that the variations were enforceable delayed the running of the limitation, and thus must surely mean that they were enforceable against both parties, so that the plaintiff would be bound to the gratuitously promised forbearance timeframe. How could it be otherwise—that a promise is binding and enforceable while it is, at the same time, unilaterally dissolvable or irrevocable? If the variation model of economic duress is to foster such artificial and uncertain interpretations, then it is hardly an improvement on consideration.

Returning again to the second rental scenario and the promise that lasted fourteen seconds, for comparison it is useful to ask what the position under the traditional approach would have been. Most assuredly, the promise would be unenforceable for want of consideration under the rule in Stilk v. Myrick. Next, with the common law working in tandem with equity, it would likely be deemed undeserving of equitable protection, since there was no reliance nor any other fairness concern that made it inequitable to return to the original contract terms after a scant fourteen seconds between promise and revocation. No doubt many would find that the landlord’s promise should hold in the first scenario covering multiple months, but should not in the second scenario, in which the tenant’s reliance on the promise could best be described as fourteen seconds of belief in a better deal. And what would explain the distinction or the different treatment between promises that are exactly the same in language and precipitating context? Likely the answer would be that the promise is unearned, or better yet, undeserved. In the second scenario, there has been no bargain, nothing contributed by the promisee tenant, nor any reliance, nor any circumstance that renders it unfair for the landlord to revert back to their original bargain. It could be that this moral or intuitive response mirrors the role played by equity and promissory estoppel. The essence of promissory estoppel is the inequity of a return. As Justice Goff confirmed in The Post Chaser, “it does not follow that in every case in which the representee has acted, or failed to act, in reliance on the representation, it will be inequitable for the representor to enforce his rights.”[75] Justice Goff further held that although the sellers of palm oil did technically rely on the buyer’s representation that delivery would still be accepted despite the seller’s delay, that the time period was so short and the extent of the reliance so insignificant that the sellers were not prejudiced in any way. Whether labelled detrimental reliance or otherwise, the key is that a representation or subsequent promise results in some changed situation between the parties such that a return to the original contract terms would be inequitable.

If the workings of equity invoke a basic fairness or moral dimension in the treatment of contractual relaxations, there is also an economic dimension. In particular, a dynamic view of efficiency can be utilized to note the problematic systemic incentives engendered by an economic duress model. A dynamic view to efficiency takes a broader view than the immediate deal, to contemplate the incentives created by legal rules. Assuming that most changes to rental agreements due to economic circumstance are practical and desirable, then it stands to reason that contract relaxations, and not only variations, should also be permitted by the law. The difficulty is that the economic duress model could have a chilling effect on such relaxations or indulgences, as parties will be tempted to insist on the original contract terms, despite practical appeal, personal inclination, or sympathy, for fear that any informal laxity will be transformed into a binding contract variation. This concern was alluded to the Supreme Court of Canada decision in John Burrows v. Subsurface Surveys by Justice Ritchie:

It is not enough to show that one party has taken advantage of indulgences granted to him … in relation to commercial transactions, such as promissory notes, it would mean that the holders of such notes would be required to insist on the very letter being enforced in all cases for fear that any indulgences granted and acted upon could be translated into a waiver of their rights to enforce the contract according to its terms. [76]

While it may be desirable to endorse socially useful contract variations, in an ironic twist this very same practical goal can be undermined in the more informal and discretionary area of contract relaxations because an overly generous approach to finding enforceable variations could easily make contract parties increasingly wary of offering relaxations at all. It could be added that parties should be wary of indulgences that become fixed contract terms, because, after all, it is not as if the promisee has done anything to deserve a new contract entitlement. Equity may well provide relief when the promisee relies on the promise, or if it would otherwise be unfair or impractical to revert back to original terms, but the change should not be formally and automatically fixed as a rewritten contract. Ultimately, when accounting for any efficiency gains from the economic duress model in permitting for useful adaptations in new circumstances, the systemic inefficiencies from the moral hazard of post-formation risk spreading and the efficiency losses from the chilling effect on relaxations and indulgences should also be discounted.

V. A Model of Changed Expectations

A.  The Problem of Promise Revocation

The primary concern with the model of economic duress is not that too many favours or indulgences will be rendered contractually binding, but rather that indulgences and gratuitous promises will be both binding and irrevocable. It is arguably an advance upon the rigid rule in Stilk to allow for promises to bind the parties even when unsupported by consideration. Yet gratuitous promises should not be resistant to revocation. Seriously intended gratuitous promises and indulgences should bind the parties in relation to their expectation of performance, but should not alter the original terms of the contract. It is suggested that the previous status quo system of consideration plus promissory estoppel implicitly reflected the wisdom that a gratuitous promise may be binding in effect through the exercise of equity, though it will not become binding as a term of the contract. The enforcement of a contract might be relaxed under equity, but the contract was not held to be rewritten. Equity could indicate that a gratuitous promise given at one’s complete discretion may not be revoked when it would unfair or inequitable to do so, but this does not suggest that the promisee has gained a contractual right beyond the control of the promisor.

A major difficulty with the economic duress model is that it is not at all open to contemplating the promisor’s position. There may be circumstances when a gratuitous promisor has miscalculated leading up to the promise, or the promisor’s position has subsequently deteriorated so that the post-formation promise becomes impractical or unsustainable. The court in Rosas sidestepped an express acknowledgement of this implication by declining to answer whether the lender could have demanded repayment of the interest-free loan before the conclusion of the implied extension period, but the implications of the economic duress model should be clear. The gratuitous promise qualified as a contract variation, thus conveniently avoiding the statutory limitation period, and contractual terms should not generally be open to alteration without the other party’s consent. That a promise that was gratuitously given should become a contractual right at the control of the promisee who did nothing in exchange for the promise or benefit is problematic on both moral and economic grounds.

No doubt some people may find it unfair that a decision to voluntarily choose to assist the other contract party who has encountered some post-formation difficulty should persist regardless of context. Again, extreme examples that help exemplify when it would be unfair to not permit for a revocation include the miscalculation corrected after only fourteen seconds after giving voice to a promise, or when the promisor is later impoverished while the promisee is enriched. Admittedly, it may not appear fair that a promisee who has agreed to provide even the slightest peppercorn of consideration would be protected from the revocation. Yet at least there is a certain formal symmetry to how the benefit or relief was granted, and that consideration reflects bilateral consent as opposed to unilateral permission. This distinction between bilateral and unilateral post-formation promises is elaborated on below.

In economic terms, it would seem inefficient to encourage a rule, as with the economic duress model, that places all of the risks of post-formation miscalculations on the promisor. Equally, as examined above in regard to the pre-existing duty rule, it is inefficient to have a rule which places too much of the post-formation risks on the promisee. A rule that placed all the risks on the post-formation promisor would no doubt incentivize greater deliberation before any promise to vary was made or any indulgence was granted. Yet carefulness is not an absolute good in terms of utility. Lord Sumption, in the recent UK Supreme Court decision in Rock Advertising, referred to the importance of informal formation in general terms: “The advantages of the common law’s flexibility about formal validity are that it enables agreements to be made quickly, informally and without the intervention of lawyers or legally drafted documents.”[77] If the utmost care is taken before any contract is formed, or any contract is altered, then there would arguably be far more cost and delay involved, which might not be worth the benefits of diligence. There could also be far fewer contracts. It may only be one time out of a hundred that a purchaser miscalculates in permitting a seller to deliver later than promised, but having an absolute prohibition on corrections (or the ability to revoke the promise unilaterally, without the seller’s consent) might be counterproductive or inefficient. A prohibition on post-formation corrections or revocations would no doubt encourage purchasers to be more diligent before all future post-formation promises, but it may mean that the purchaser expends far more resources or effort before making a promise, and as a result winds up entering into fewer beneficial promises. The costs of increased diligence—to the parties in the actual diligence taken or to society in the amount of beneficial commercial promises entered into—is not socially useful if most of the time taking less care yields the desired outcome at far less cost. A model based on the changed expectation interests of the parties is argued to provide a certain formal symmetry in how a post-formation promises are made, and provide a balanced perspective on risk bearing between the parties.

B.  Change to Contract Expectations

Why should a gratuitous promise to alter future contract performance be binding? While there are various ways to justify the enforcement of promises, such as the expression of individual autonomy, under a model of changed expectations the justification is simply based on the legitimate expectations of the parties. “Legitimate expectations of the parties” has become something of a watchword for justifying the reform of Canadian contract law, as both the decisions in Nav Canada[78] and Rosas,[79] as well as the Supreme Court’s landmark decision in Bhasin,[80] demonstrate. As applied here, the legitimate expectations of the parties do not simply denote a general sense of what parties may anticipate in the nature of standard business practices, as in Bhasin, but also include the circumstances and plans specific to the contract parties at hand.

A model of changed expectations would be based on the following basic propositions: 1) a post-formation promise should be binding because it alters the expectations of both contract parties; 2) performance is bound by the expectations of the parties at the time that it is tendered; 3) the post-formation expectations of the parties may be altered in the same manner they were set by the parties. If there is anything notable in the above propositions it is that all promises may equally set the expectations of the parties, whether supported by consideration or not, which will govern performance subject only to revocation and its timing. The key element, then, is revocation. A gratuitous promise may be revoked unilaterally, a bilateral promise supported by consideration cannot be.

Remaining alive to the desirability of permitting for some possibility of revocation of gratuitous promises establishes a priority on variability, and emphasizes the importance of differing levels of enforceability. Bargained-for promises, those supported by consideration, are binding and are irrevocable without the consent of the other party. Gratuitous promises should be both binding and revocable. Or, put differently, voluntary promises are binding until the point of unilateral revocation (with reasonable notice). When the promisor voluntarily promises a gratuitous benefit, this can be thought of as unilateral in the sense that control of the alteration was exercised by only one of the contract parties and did not depend on the other’s consent. Clearly, a beneficiary will often have to consent to receive the benefit, as in receiving additional funds or providing late delivery, but the use of unilateral here is simply meant to denote a one-sided commitment in which the other party does not promise to do or provide anything new or different.[81] When a variation is supported by consideration and there is a reciprocal exchange of promises, “I agree to accept less if you agree to pay early,” it can be deemed bilateral. A gratuitous promise, since it is without any commitment from the beneficiary, should be under the promisor’s sole control in a way that bilateral variations or changes are not; gratuitous promises should only be subject to the limit of reasonable notice or past performance. No doubt courts could infuse some concerns of equity or fairness into the assessment of what constituted a reasonable time for notice.

How the legitimate expectations of the parties are set should determine how they are subsequently altered. If the variation occurred with the other party’s consent in a bilateral exchange, this should require the other party’s consent to revert to the original terms. When the promise or indulgence was gratuitous, and thus unilateral, the promise can be unilaterally revoked. The major caveat to the unilateral revocation would be the requirement of reasonable notice. A reasonable person must foresee that a contract party’s performance cannot, as a practical matter, normally be instantaneously rearranged. The gratuitous promisee should not have an irrevocable contractual right to the promise, but they should be granted a reasonable amount of time to arrange their affairs in order to return to the original expectations of performance set out in the contract.

Significantly, a gratuitous promise can be unilaterally revoked in principle, but not always in practice. The gratuitous permission to receive late delivery of a machine, for instance on March 15th instead of the original agreed upon date of February 1st, cannot be revoked on March 30th after the machine has been delivered and accepted by the purchaser. This resembles the reasonable notice operation under promissory estoppel.[82] Unlike with estoppel, however, under a model of changed expectations there would be no need to inquire into detrimental reliance, nor would it depend upon any sword-versus-shield distinction. This provides for a more efficient or streamlined approach to the protection of reliance than under existing doctrines. While a party’s reliance on a promised variation or relaxation may explain why a variation is deserving of protection, the demonstration of this reliance has become cumbersome and often artificial. A model of changed expectations would forgo the reliance requirement, and the qualitative inquiry it entails, to instead simply enforce the variation because it was seriously intended by the promisor. The equitable evaluation of reliance on the part of the promisee is switched out for a mere serious intention on the part of promisor. A promised variation is enforceable because it was a promise seriously intended, nothing more. However, the ability to return to original terms is preserved when there was no exchange of promises for the variation, or consideration. When nothing was given in exchange by the promisee, then the promised variation should be revocable by the promisor, subject to reasonable notice. The reasonable notice requirement will invariably protect the reasonable reliance on the part of the promisee, but in a much more direct and straightforward manner than under the doctrines of promissory or equitable estoppel.

Under a model of changed expectations, performance is bound by the expectations of the parties at the time that it is tendered, and therefore the timing of the revocation is decisive as to the enforceability of a gratuitous promise. The yes-or-no quality of previous models—in which the pre-existing duty rule would categorically say no to a gratuitous promise (so that judges strive creatively to “find” consideration), and the economic duress approach that will always answer yes—is arguably averted under a model of changed expectations.[83] If the first question on the enforceability of contractual changes involves the possibility of revocation—was the promise bilateral or unilateral?—the second question would ask how much performance occurred under the promise before revocation and the reasonable time for a return. A model of changed expectations would revolve around the notion of until revocation—a gratuitous promise that changes expectations is binding until revocation with reasonable notice.

The potential advantage of changed expectations over both consideration plus estoppel and the model of economic duress can be further illustrated with two brief examples. First, suppose an individual buys a ticket to a popular musical production that is to take place on Monday night. Unfortunately, the theatre owner has overbooked seats for Monday’s performance. The theatre owner contacts the guest, explains the situation and requests that the guest attend the same show in the same numbered seats on Tuesday night instead. The owner does not offer any incentives or discounts to the guest. The guest lives across the country and will only be in town on Monday and Tuesday. The guest is agreeable and expressly consents to change the tickets from Monday to Tuesday. At this point the theatre owner has a changed expectation interest, but it is a notional interest only. The owner has not acted on the promise, nor provided anything for the benefit. To clarify, suppose that Monday day tickets are worth one dollar more than Tuesday tickets and that the owner is free to sell the Monday ticket to another guest. At this point would it be correct to say that the owner is one dollar wealthier? Not necessarily, because it is notional only. The owner may well realize the additional profit from the higher Monday matinee ticket price if there is sufficient demand, but no sale has been secured yet and one may not materialize. Notional interests are insecure, and should not have the value of legal entitlements that were bought or bargained for. In short, the economic value of a gift that is promised to us and the value of a gift that has been delivered into our hands is very different, because the former is an uncertain, revocable expectation. Shortly after agreeing to the change, the guest consults her calendar and recalls that she already has a serious commitment on Tuesday night that cannot be cancelled. The guest apologetically tells the owner that she cannot attend on Tuesday and would like her Monday tickets back. The owner has not acted upon the ticket change in any way. Should the guest be bound to the non-refundable Tuesday ticket? Only the model of economic duress model would hold the guest to the change. Under a model of changed expectations, the guest could unilaterally revoke her gratuitous promise because she gave the theatre owner sufficient notice of the return to the original contract expectations. The theatre owner had no expectations or actions to unravel after only a moment. In effect, the guest’s speedy notice served to prevent any reliance on the part of the theatre owner. The guest would likely not have much time to revoke in this circumstance, however. In many single round contracts such as this one, with only one delivery of a good or service and without subsequent stages of performance, once the promisee has entered into performance it would be too late to revoke with reasonable notice. In this example, once the theatre owner has assigned the Monday ticket to another guest to correct the booking error, it would be too late to provide reasonable notice. This is so because it would be unreasonable to require the theatre owner to rescind the ticket assignment to the other guest. Needless to say, once the Monday performance has taken place, the guest could not then revoke and expect to attend a performance in the past. Additionally, this would be an optimal revocation of a promise because the theatre owner is arguably better placed to find a solution than the guest, and there is no substantial cost to the theatre owner in the guest’s fleeting miscalculation. Unlike the model of economic duress, the consideration plus estoppel status quo approach would provide a straightforward answer to the above example, though with two further steps. Again, those two further steps under the promissory estoppel approach would involve the question of whether the return was inequitable due to detrimental reliance or not, and whether the guest would happen to be the plaintiff (sword) or the defendant (shield).[84] Perhaps these two further steps would not involve much time or cognitive difficulty in this particular case, but it is important for comparison to recall that these steps are present, and that they might not always be disposed of so easily. It is in the next example that the consideration plus equity approach arguably falters.

Imagine circumstances nearly identical to those found in Stilk, in which a ship’s captain voluntarily approaches the ship’s crew with a promise of increased wages after some early desertions. Assume instead, however, that under the original contract the voyage is divided into two instalments, in which each crew member is to receive five dollars for the outward leg from London to the Balkans and five dollars for the return leg, for a total of ten dollars. Under the contract, the ship owners pay for all of the crew’s supplies and expenses while at sea and while away at the destination port. Let us further suppose that the captain’s promise to pay increased wages, at a rate of seven dollars per leg of the journey, was made before the first leg of the voyage while the crew waited to set sail in London. Further assume that upon arriving in the Balkans, the captain discovers that the goods awaiting shipment have deteriorated and are of far less value than anticipated. The project is not frustrated, but the voyage will be a near total loss for the shipowners.

While in the Balkan port waiting for the inferior cargo to be loaded, the captain informs the crew that they will not be paid any increased wages for either leg of the voyage. The crew is disgruntled but require passage back home to London, and there are no other ships in port. The crew eventually boards the ship and sails it safely back to London. Subsequently, the shipowners pay each crew member wages of ten dollars, but refuse anything more. The sailors insist they ought to be paid fourteen dollars for the journey based on the captain’s promise. Following Nav Canada and Rosas, in the absence of duress, the captain’s promise would count as a valid contract variation for an outcome of fourteen dollars for the sailors. Following the classic approach from Stilk, the captain’s promise of seven dollars was gratuitous and would be unenforceable, for an outcome of ten dollars for the sailors. But what of promissory estoppel?

It is suggested that an equitable approach to the hypothetical would either be extremely convoluted on the one hand, or superficially straightforward and unsatisfying on the other. For one, it is conceivable that the application of promissory estoppel would become bogged down in an evaluation of reliance and the equities between the parties. One could follow Lord Denning’s lead and find that the sailors acted on the captain’s promise and that it would be inequitable to retract the promise. In contrast, this scenario would fail under the traditional estoppel requirement for detrimental reliance, for if reliance is to constitute something other than merely performing what one was already bound to do, the sailors have not altered their position in any way based on the captain’s promise. Additionally, the position of the sailors has not improved, for no more crew members were added to compensate for the early desertion, and so the implied expiration found in High Trees would not apply here. Instead, the change in circumstances is entirely on the side of the promisor, and this at least suggests that a return may not be inequitable. This is not to suggest that there would not be an answer available under equity and promissory estoppel, only that it might be muddled or incoherent. It should also be stressed that this mild criticism of promissory estoppel is intended to be in general, because in this specific hypothetical there would be a clear-cut, and rather unsatisfying, answer: the sailors would need to be plaintiffs in order to recover from the defendant shipowners, and the use of promissory estoppel as a sword would be categorically barred. Thus, the outcome under promissory estoppel would likewise be a total of ten dollars.

Finally, how would the hypothetical be treated under a model of changed expectations? The captain’s gratuitous promise revised the expectations of both parties, yet it could be revoked unilaterally subject to reasonable notice. How long would the crew members reasonably need to arrange their affairs back in line with the expectations set in the original contract? For the sake of argument let us assume that the period required for reasonable notice would be quite short and would, at the very latest, run until the second leg of the voyage began. The crew’s expenses are covered by the shipowners at sea and in port, and they are bound, by contract and by practical necessity, to sail with the ship back to London. Unlike the theatre example, there is no predictable material change or investment that the promisee crew members would undertake in reliance on the gratuitous promise. The crew would be operating under a purely notional financial expectation, and as such this expectation could be revoked unilaterally, at least until the second leg of the voyage commences. Revocation cannot reasonably occur when it is no longer feasible, or when there is no remaining performance that is severable into different expectations or price points. As a leg of the voyage is unlikely to be severable on either a per diem or distance traveled basis, one rate is likely to cover the return leg as a single unit of service. For example, the crew would unlikely be expected to return wages if favourable winds made for a shorter voyage than expected. Put a different way, in a contract for the delivery of flour, the delivery en route under a gratuitous promise is irrevocable, but the delivery for a following week is revocable. Practicality as much as fairness yields the result, to avoid the absurdity of attempting to assign damages based on flour that leaves the mill at seven dollars but is revoked for an original contract price of five dollars while the delivery truck is in transit. (If the prospect of the crew abandoned in the Balkans seems harsh, one might imagine that the contract involved the crew of a local ferry on the Thames, in two installments months apart.) The first leg of the journey would have operated under the captain’s gratuitous promise. The sailor’s performance for the first leg is already past when the unilateral revocation was made. The captain’s promise was gratuitous and did not irrevocably vary the contract, but it was seriously intended and it governed the expectations of both parties during the crew’s performance during the first leg. The second leg of the voyage occurred after revocation with reasonable notice, and so it was governed by the original contract terms. A view to changed expectations would yield a total result of twelve dollars for the crew.

The above examples were crafted for the purpose of demonstrating that there are important differences between the various models of contract variation. It matters whether consideration or economic duress is the standard for evaluating contract enforcement, and it matters whether promissory estoppel is applicable as a matter of either substance or procedure. In comparison, it is suggested that a model of changed expectations would be more efficient and easier to apply than the alternatives, while promoting fair results.

Conclusion

It is no doubt worthwhile to permit gratuitous contract changes that respond to new economic realities, but this should not come at the expense of undermining the potential for unilateral contract relaxations. It is suggested that a model of changed expectations provides a straightforward and efficient framework, which connects normative justification with practical application. A model of changed expectations provides a simple relationship between the why, how, and what of contact variation enforceability. Why is a post-formation promise binding? Because the promise altered the legitimate expectations of the parties. How does the promise change expectations? The promise could change expectations unilaterally, through one party’s gratuitous favour, or bilaterally through the exchange of an additional bargain. How can expectations be changed back to the original contract terms? In the same manner as the variation or relaxation: either unilaterally or bilaterally, meaning with or without the other contractual party’s permission. What is the enforceable outcome? The amount of performance contributed or tendered under a set of expectations, whether supported by consideration or not, is recoverable at the rate of the altered expectation.

Criticisms of the pre-existing rule are well-founded, but the solution in the model of economic duress may be a blunt oversimplification that corrects one shortcoming but produces another. The major flaw of the economic duress model is that it supplants the role of promissory estoppel, and the equitable balancing that it provides. The model of changed expectations, by contrast, may combine the advantage of endorsing seriously intended variations with the fairness and flexibility of equity. A model of changed expectations promotes a balance between favour and entitlement, and between expectations and fairness.

                                   

[1]     Combe v Combe, [1952] EWCA Civ 7, [1951] 2 KB 215 at 220.

[2]     See John D McCamus, The Law of Contracts, 1st ed (Toronto: Irwin Law, 2005) at 381–82.

[3]     See e.g. BJ Reiter, “Courts, Consideration, and Common Sense” (1977) 27:4 UTLJ 439 [Reiter, “Common Sense”] (“Legal rules should be retained only so long as they promote desirable social policies, only so long as they are consistent with sense and reason, and only so long as they help courts make decisions. Judged by any of these criteria, the pre-existing duty rule must go” at 506–507); Angela Swan, Jakub Adamski & Annie Y Na, Canadian Contract Law, 4th ed (Markham: LexisNexis, 2018) (commenting on contemporary case law undermining the pre-existing duty rule: “If the result of this development were that, as has been suggested, all modifying arrangements or undertakings made in the context of a commercial relation were to be enforced (absent some real reason not to) that would be a significant improvement over the existing situation” at § 2.204).

[4]     [1989] EWCA Civ 5, [1991] 1 QB 1 [Roffey Bros].

[5]     [1809] EWHC KB J58 [Stilk].

[6]     On the judicial developments in New Zealand and Canada see Part 4, below; Courts in the United States, under the influence of the Restatements, long ago relaxed the consideration requirement.

[7]     Under equity, the stipulation that promissory estoppel was only available as a shield and not a sword effectively served to distinguish between claims for more (sword) from promises to pay less (shield). At common law, promises to pay more or less were prescribed a different treatment in a Court of Appeal decision which restricted the application of Roffey Bros to promises to pay more (Re Selectmove Ltd, [1993] EWCA Civ 8). See also MWB Business Exchange Centres Ltd v Rock Advertising Ltd, [2016] EWCA Civ 553 [Rock Advertising 2016]; Rock Advertising Ltd v MWB Business Exchange Centres Ltd, [2018] UKSC 24 [Rock Advertising 2018].

[8]     See e.g. Judicature Act, RSA 2000, c J-2, s 13(1).

[9]     [1875] LR 10 Ex 153 at 162, CCS 34.

[10]    See Stilk, supra note 5.

[11]    See ibid.

[12]    See Alaska Packers’ Ass’n v Domenico, 117 Fed 99 (9th Cr 1902) [Alaska Packers’].

[13]    (1791) Peake 170 ER 94, [1775–1802] All ER Rep 493.

[14]    See ibid at para 103.

[15]    On the conflicting reports of the case, see generally Peter Luther, “Campbell, Espinasse and the Sailors: Text and Context in the Common Law” (1999) 19:4 LS 526.

[16]    Stilk, supra note 5.

[17]    Richard A Posner, “Gratuitous Promises in Economics and Law” (1977) 6:2 J Leg Stud 411 at 422.

[18]    See e.g. Charles Fried, Contract as Promise: A Theory of Contractual Obligations (Cambridge, Mass: Harvard University Press, 1981) at 1–17, 28–39.

[19]    See e.g. PS Atiyah, The Rise and Fall of Freedom of Contract (Oxford, UK: Oxford University Press, 1979) at 398–454, 716–79; Grant Gilmore, The Death of Contract (Columbus: Ohio State University Press, 1974) at 5–35, 87–103; Barry J Reiter, “Contracts, Torts, Relations and Reliance” in Barry J Reiter & John Swan, eds, Studies in Contract Law (Toronto: Butterworth, 1980) 235.

[20]    See e.g. John Swan, “Consideration and the Reasons for Enforcing Contracts” in Barry J Reiter & John Swan, eds, Studies in Contract Law (Toronto: Butterworth, 1980) 23 at 24–40; Reiter, “Common Sense”, supra note 3.

[21]    As opposed to statutory steps, such as the Judicature Act, supra note 8, and the American restatement project.

[22]    Roffey Bros, supra note 4 at 16, Glidewell LJ.

[23]    Ibid at 13–14.

[24]    See Pao v Lau, [1979] UKPC 17 at 9–10, 13.

[25]    See Roffey Bros, supra note 4 at 13–15.

[26]    See e.g. Rock Advertising 2016, supra note 7. But see Rock Advertising 2018, supra note 7.

[27]    See Musumeci v Winadell Pry Ltd, (1994) 34 NSWLR 723; Marcus Roberts, “Variation Contracts in Australia and New Zealand: Wither Consideration?” (2017) 17:2 OUCLJ 238 at 241–44 [Roberts, “Variation Contracts”].

[28]    See Brian Coote, “Consideration and Benefit in Fact and in Law” in Rick Bigwood, ed, Contract as Assumption: Essays on a Theme (Oxford, UK: Hart Publishing, 2010) 53 at 56–59.

[29]    See ibid at 57–59; Rick Bigwood, “Doctrinal Reform and Post-Contractual Modifications in New Brunswick: NAV Canada v. Greater Fredricton [sic] Airport Authority Inc.” (2010) 49:2 Can Bus LJ 256 at 267; MH Ogilvie, “Of What Practical Benefit Is Practical Benefit to Consideration?” (2011) 62 UNBLJ 131 at 135–46.

[30]    See Roberts, “Variation Contracts”, supra note 27 at 244; South Caribbean Trading Ltd v Trafigura Beheer BV, [2004] EWHC 2676 (Comm) at 108. See also Adam Opel GmbH v Mitras Automotive UK Ltd, [2007] EWHC 3481 (QB) at paras 40–42.

[31]    [2003] 2 NZLR 23 (NZCA) [Antons].

[32]    Ibid at para 93.

[33]    Ross Bindon Ltd v PB & CS Properties Ltd (in liqu), [2006] 7 NZCPR 850 (HC) at
para 36.

[34]    Roberts, “Variation Contracts”, supra note 27 at 252–54.

[35]    [2008] 290 DLR (4th) 405, 329 NBR (2d) 328 [Nav Canada].

[36]    Ibid at para 27.

[37]    Ibid at paras 31–32.

[38]    See Kennedy v Clark, [2009] 313 DLR (4th) 738, 349 NBR (2d) 91.

[39]    See Globex Foreign Exchange Corp v Kelcher, [2011] 337 DLR (4th) 207, 48 Alta LR (5th) 215.

[40]    2018 BCCA 191 [Rosas].

[41]    Ibid at para 183.

[42]    See Antons, supra note 31 at para 93.

[43]    Rosas, supra note 40 at para 165.

[44]    Ibid at paras 174–76.

[45]    Ibid at para 183.

[46]    Marcus Roberts, “Rosas v Toca: Asses and ‘Incremental Changes’ to Consideration” (2019) 61:3 Can Bus LJ 392 at 401. See also John D McCamus, The Law of Contracts, 2nd ed (Toronto: Irwin Law, 2012) at 254.

[47]    See Posner, supra note 17 (Posner begins with three categories, but distills them down to two at 422–23).

[48]    See Varouj A Aivazian, Michael J Trebilcock & Michael Penny, “The Law of Contract Modifications: The Uncertain Quest for a Benchmark of Enforceability” (1984) 22:2 Osgoode Hall LJ 173 at 174.

[49]    See Posner, supra note 17 at 422.

[50]    See Alaska Packers’, supra note 12.

[51]   See ibid at 102.

[52]    47 Mich 489, 11 NW 284 (1882).

[53]    Ibid at 492–93.

[54]    See Posner, supra note 17 at 421–22.

[55]    See ibid at 422–23.

[56]    See Aivazian et al, supra note 48 at 184–87. See also Andrew Phang, “Whither Economic Duress? Reflections on Two Recent Cases” (1990) 53:1 Mod L Rev 107.

[57]    See Aivazian et al, supra note 48 at 195.

[58]    See generally Lorne Neudorf & Geoffrey Hunnisett, “Force Majeure Clauses in Comparative Perspective: The Canadian Common Law Approach in Light of Recent Developments in the Courts of Singapore and the United Kingdom” (2014) 65 UNBLJ 312. See also Michael P Theroux & April D Grosse, “Force Majeure in Canadian Law” (2011) 49:2 Alta L Rev 397 .

[59]    This is an area where doctrines of frustration and mistake overlap.

[60]    See Aivazian et al, supra note 48 at 175.

[61]    See ibid.

[62]    On private ordering, see Steven L Schwarcz, “Private Ordering of Public Markets: The Rating Agency Paradox” (2002) 97:1 U Ill L Rev 1. See also Tehila Sagy, “What’s So Private about Private Ordering?” (2011) 45:4 Law & Soc’y Rev 923; Barak D Richman, “Firms, Courts, and Reputation Mechanisms: Towards a Positive Theory of Private Ordering” (2004) 104:8 Colum L R 2328.

[63]    Aivazian et al, supra note 48 at 211–12.

[64]    See Rosas, supra note 40.

[65]    See ibid at para 2.

[66]    See ibid at paras 2, 15–16.

[67]    See ibid at para 2.

[68]    See ibid at para 17.

[69]    See ibid at paras 177, 179: although the court opted not to follow Williams v Roffey, it observed that, “if one were driven to find consideration” there were practical benefits that flowed to the plaintiff, such as maintaining the friendship with defendant. Fortunately, the court decided to resist such “machinations.”

[70]    Ibid at para 1.

[71]    The plaintiff dropped an estoppel claim at trial and it was not reconsidered on appeal (ibid at paras 44–48).

[72]    Central London Property Trust Ltd v High Trees House Ltd, [1946] EWHC KB 1, [1956] 1 All ER 256 at 259.

[73]    Cf Antons, supra note at 31 at 59, where the court’s reasons admittedly included language of the promisee “acting on” the gratuitous promise. This would avoid the problem of enforceability for the fourteen second promise. While this is preferable to an automatic contract variation permitted by Nav Canada, supra note 35 and Rosas, supra note 40, it still does not provide the potential for a unilateral return. That is, once the tenant relies upon the promise, however little, it becomes a contract variation, and the landlord would not be able to initiate a return, regardless of notice or what is fair in the circumstances.

[74]    Rosas, supra note 40 at para 29.

[75]    Societe Italo-Belge Pour Le Commerce et L’Industrie SA v Palm & Vegetable Oils (Malaysia) SDN BHD (sub nom The Post Chaser), [1982] 1 All ER 19 (QB) at 27 [The Post Chaser].

[76]    John Burrows Ltd v Subsurface Surveys Ltd, [1968] SCR 607 at 615, 68 DLR (2d) 354.

[77]    Rock Advertising 2018, supra note 7 at para 12.

[78]    See Nav Canada, supra note 35 at para 28.

[79]    See Rosas, supra note 40 at paras 4, 28, 165–67, 176.

[80]    See Bhasin v Hrynew, 2014 SCC 71.

[81]    Unilateral as used here is not intended in the sense of a unilateral contract offer, which others have used to provide a theory of contract variation. See Mindy Chen-Wishart, “Consideration for Variation of Contracts” [2014] NZLJ 67; Mindy Chen-Wishart, “A Bird in the Hand: Consideration and Contract Modifications” in Andrew Burrows & Edwin Peel, eds, Contract Formation and Parties (Oxford, UK: Oxford University Press, 2010) 89; Mindy Chen-Wishart, “Consideration and Serious Intention” (2009) 2 Sing JLS 434 at 439–41. For criticisms of this unilateral offer approach, see Coote, supra note 28 at 58; Roberts, “Variation Contracts”, supra note 27 at 248.

[82]    See e.g. International Knitwear Architects Inc v Kabob Investments Ltd, [1995] 17 BCLR (3d) 125, 67 BCAC 128.

[83]    As to the issue of duress, there are two potential caveats to note: 1) coercion by the promisee could be said to defeat the expectation view of the promise in both the expectation of the promise and the promisor; and 2) the coerced promisor’s resistance should serve as notice that the original contract terms are back in force, or were never altered legally.

[84]    The simple fact of whether the guest had already paid for the tickets or not would likely be central to resolving this issue.