Certainty, Justice and the Law of Agency in the Chinese Civil Code: A View from England

Thomas Krebs＊

University of Oxford

Abstract: The new Chinese rules on agency do not impose broad “fiduciary”duties on agents—instead, there are a number of specific provisions designed to protect the principal against particular abuses to which it is peculiarly vulnerable in the principal/agent relationship. Chinese law, thus, deliberately refuses to follow the lead of English law, which imposes very strict and wide-ranging fiduciary duties on agents. This paper argues that this is probably wise. English law has to be seen against a matrix of a system of commercial law which was forged on the anvil of international trade and commodity supply contracts, leading to a set of rules that prefer certainty of outcomes (and the avoidance of litigation) overachieving particular justice in individual cases (such as might have been achieved by subjecting English law to an overarching“good faith” principle). English commercial law is adversarial, not cooperative. This explains why, in a relationship that is characterized by cooperation, such as the principal/agent relationship, the general rules of English commercial law are replaced by wide, justice-oriented rules.A system that is already based on cooperation, for which Chinese law is almost paradigmatic, is likely much more adept at applying the general rules to the agency relationship than English law would be.

Keywords: agency, English commercial law, agent’s duties to principal, fiduciary duties

Introduction

English commercial law is characterized by its uncompromising and unapologetic adversarial nature. It is not necessarily designed to promote justice but certainty.①Cf. Lord Mansfield’s famous dictum in Vallejo ν Wheeler (1774) 1 Cowp 143 at 153: “in all mercantile transactions the great object should be certainty: and therefore, it is of more consequence that a rule should be certain, than whether the rule is established one way or the other. Because speculators in trade then know what ground to go upon.”Commercial parties need to know where they stand, and for that, they need clear rules and bright lines, reducing the incidence of judicial discretion to a bare minimum. Promoting certainty reduces litigation, and the theory goes: If the parties know what the rules are,they or their legal advisers can apply them to the problem at hand and accurately assess their chances in court. This emphasis on certainty is one reason, as is often claimed, why“English commercial law is chosen around the world by commercial counterparties to govern their contracts, even when neither they nor the subject matter has any connection with England.”②Lord Briggs in Manchester Ship Canal Co Ltd ν Vauxhall Motors Ltd [2019] UKSC 46, [2020] 2 All ER 81, [2019] 3 WLR 852, at [41]. This is, of course,not uncontroversial. There are many reasons, other than its supposed superior quality and the expertise of those who apply it, why English law may have the popularity it undoubtedly possesses. Language is, I would argue, the most important of these. While Mandarin Chinese is the language with the most native speakers (English only makes the third place, after Spanish, in that list), it is by a wide margin the most commonly spoken second language in the world:“Summary by language size.” Ethnologue. Retrieved 12 March 2019.A function of the focus on certainty is a rejection of a general principle of good faith: English lawyers claim that they do not know what “good faith” means③Cf. Ewan McKendrick, Goode and McKendrick on Commercial Law, LexisNexis (London), para 3.75 (2020).and fear that it will serve to inject discretionary justice into commercial law, thus undermining the very foundation of its success. Commercial law, and indeed contract law more broadly, is regarded, by the English lawyer, as adversarial in nature, each party having to look after itself, and this concern has been held to render even a contractual agreement by which the parties purported to bind themselves to good faith negotiations void for uncertainty.④Walford ν Miles [1992] 2 AC 128.

Legal systems that locate themselves in the Civilian tradition are committed to a generally more collaborative approach. Contracts are not seen as contests, not as competition, but as relationships that impose an imperative on both parties to work out a solution to their differences. This is, of course, particularly pronounced in the Chinese tradition going back millennia: Disputes are better resolved by negotiation and mutual understanding, if need be, by mediation; going to court is considered an evil in itself and certainly a last resort.⑤Cf. Konrad Zweigert and Hein K?tz, Einführung in die Rechtsνergleichung auf dem Gebiete des Priνatrechts, Mohr (Tübingen), at 282-286 (1996).The goal—avoiding litigation and the social strife that this brings—is the same, but the approach is fundamentally different.

This article considers some of the provisions relating to the law of agency in the new Chinese Civil Code (Ciνil Code of the People’s Republic of China, hereinafter theCiνilCode) from an English perspective. Agency is an interesting area of the law because its division into internal and external relationships lays bare a fundamental tension in the English “style” of commercial law: externally, i.e., in the context of the principal—third party relationship, the same (harsh) rules obtain that characterize English commercial law generally (it is seen as adversarial), the internal relationship between principal and agent is recognized as collaborative. Principal and agent are “on the same team,” and therefore,need to be able to trust one another. English law has long recognized that, by giving an agent authority to bind him, a principal is, to some extent, at the mercy of the agent. If the relationship was an adversarial one, this simply would not work: If the agent was “in it for himself,” he would not be able to do what he was appointed to do, namely, to represent the principal and to protect the principal’s best interests.①It is remarkable, but not surprising, that English law did not, by itself, appreciate that, apart from the principal needing protection against exploitation by the agent, the converse was also the case. Thus, an agent who had built up a distribution network, goodwill, or a customer base on behalf of the principal had no recourse if the principal, at this point, decided to dismiss it while taking advantage of the fruits of the agent’s labors. It took the intervention of EU law, in the form of Directive 86/653/EEC (implemented in the UK by the Commercial Agents (Council Directive) Regulations 1993, so far surviving the departure of the UK from the European Union), to remedy this.English law is conspicuously emphatic when it comes to signaling this change in emphasis: The agent is made subject to “fiduciary obligations” that are so strict that one might call them draconian in nature.They certainly go well beyond the more specific exhortations that can be found in Arts.164 and 168 of the ChineseCiνil Code.

The underlying argument made in this article is that it is not at all uncommon for English law to temper its “bright line” approach focusing on certainty rather than equity by going to the other extreme where the “bright line” approach simply fails to work or manifestly produces injustice. A system of private law which, from the start, is less adversarial and more cooperative in nature and style may thus be well advised not to emulate the English approach.

Certainty v Justice in English Private Law

English commercial law, and indeed much of English private law generally, has been developed in the context of international trade and commerce and can only be understood against that background. English law students will quickly become familiar with case names that are simply the names of ships—these are some of the most important cases that they will have to study and learn. Thus, the rule that pure economic losses cannot be recovered in a tort action was first authoritatively laid down by the House of Lords, then the highest court in the land, in The Aliakmon;②Leigh & Sillaνan Ltd ν Aliakmon Shipping Co Ltd [1986] AC 785. The House of Lords approved and applied another shipping case, decided by the Commercial Court, in The Wear Breeze [1969] 1 QB 219.the Court of Appeal in The Hong Kong Fir③Hong Kong Fir Shipping Co Ltd ν Kawasaki Kisen Kaisha Ltd [1962] 2 QB 26.laid down a typology of contractual terms, adding the category of “innominate”terms to the well-established “conditions” and “warranties”.④A breach of condition entitles the innocent party to suspend its own performance and terminate the contract from that point onwards, whereas an action for breach of warranty sounds in damages only—ironically, the shipping case Hong Kong Fir obscured that bright line, otherwise so typical for English commercial law, by introducing a third category of term, where the consequences of its breach determine whether the innocent party is entitled to terminate:the more serious they are, the more likely the innocent party is entitled to walk away. Gill & Duffus SA ν Berger & Co Inc (No 2) [1984] AC 382.The Albazero,⑤Albacruz (Cargo Owners) ν Albazero (Owners) [1977] AC 774.affirming,though distinguishing, another shipping case,Dunlop ν Lambert,⑥D(zhuǎn)unlop ν Lambert (1839) 6 Cl & F 600.deciding more than a century previously that a contracting party may, in some circumstances, recover losses incurred not by itself but by the transferee of the property affected. The “rule in the Albazero” was subsequently applied in a long line of construction cases involving defective building work, starting with the House of Lords decision in Linden Gardens①Linden Gardens Trust Ltd ν Lenesta Sludge Disposals Ltd [1994] 1 AC 85.and culminating in the controversial decision, by the same court, inAlfred McAlpine Construction Ltd ν Panatown Ltd,②Alfred McAlpine Construction Ltd ν Panatown Ltd [2001] 1 AC 518.which is required reading for any contract law student given its fundamental significance for the meaning of loss in English contract law. The original problem inDunlop ν Lambertwas caused by the strict English rule that only a party to a contract can sue on it, the rule of “privity,” a rule that proved to be a particular nuisance in shipping cases, so that the leading cases on privity arose, again, in the context of international trade.③Scruttons ν Midland Silicones [1962] AC 446; New Zealand Shipping Co Ltd ν Satterthwaite, The Eurymedon [1975] AC 154; The Mahkutai [1996] 3 WLR 1.It is in the nature of international trade in general, and the trade in commodities in particular, that “time is money” and that the parties value certainty above everything. The premium that is put on certainty can lead to results that are, to the nonspecialist, surprising and counterintuitive. Thus, in the case ofGill & Duffus SA ν Berger& Co Inc (No 2),④the House of Lords held that even where it was clear that beans that had been bought under a c.i.f. contract were of the wrong quantity (which would ordinarily justify rejecting them), the buyer had to pay against conforming documents. This makes sense within the context of the c.i.f. contract, but the point I am trying to make here is that frequently English commercial law, and indeed English contract law, seems designed to facilitate the commodity trader at the expense of the small business owner or consumer,and this is because its rules were hammered out on the anvil of international trade law.

The emphasis on certainty is pervasive in the core statute in English commercial law,the Sale of Goods Act 1979, and in the way in which its provisions have been interpreted. S.35 SGA may serve as an example. As already mentioned, a breach of condition entitles the innocent party to terminate the contract. In the context of the sale of goods, the innocent buyer is, upon the seller’s breach of condition, entitled to reject the goods. The most obvious condition that might be breached by a seller is implied into the sale contract by s.14(2) of the Act, to the effect that the goods must be of satisfactory quality. This is widely defined in the section—even slight breaches—slight defects—entitle the buyer to reject and receive a full refund of the price. Rather than requiring a buyer who chooses to reject having used the goods for a time to account for the use value made of the goods, following the controversial case ofRowland ν Diνall,⑤Rowland ν Diνall [1923] 2 KB 500.English law simply ignores this issue.Instead, s. 35 SGA temporarily restricts the right to reject, giving the buyer only a very small time window within which the remedy of rejection can be exercised.⑥The high point of this strict approach was reached in Bernstein ν Pamson Motors Ltd [1987] 2 All ER 220, in which Rougier J held, albeit only at first instance,that use of a car over a three-week period barred the remedy of rejection when a defect was discovered. It has since been somewhat relaxed: Truk Ltd ν Tokmakidis [2000] 1 Lloyds Rep 543; Clegg ν Olle Anderson [2003] 2 Lloyds Rep 32.Commercial law prefers certainty over justice—it would rather not get involved with the tedious and expensive inquiry into how much the buyer’s use of the goods over a period of time would be worth in the market.①It is ironic and makes my point rather well that, for consumers, to whom the Sale of Goods Act 1979 applied, subject to the Unfair Contracts Terms Act 1977,in its full rigor until then, English law was changed in 1994 following an EU Council Directive (93/13/EEC, now superseded by Directive 1999/44/EC),leading to much more nuanced, but much less certain, provisions of the Consumer Rights Act 2015 in ss. 19-24, giving the consumer buyer the option to have the goods repaired or replaced, or to reject them against a refund reduced by an appropriate amount to reflect the use made of them. The Directive, of course,was very much influenced by the civilian tradition, in particular by German law.

The preference of English law for certainty over justice, at least up to a certain point,can also be demonstrated by examining the foundational rule of English contract law:freedom of contract. It is one of the most basic tenets of English contract law that the courts will not inquire into the adequacy of the consideration. In other words, they will not adjust the parties’ bargain so as to make it “fair.”②See, for many, Gaumont-British Pictures Corp ν Alexander [1936] 2 All ER 1686; Langdale ν Danby [1982] 1 WLR 1123; Brady ν Brady [1989] AC 755.This general rule is even left untouched by consumer protection legislation, in that the Consumer Rights Act 2015, in s.64, excludes “core” terms of the contract, i.e., those that determine the subject matter of the contract and the price, from the general assessment of fairness that all other terms are subject to. Yet the unconcern with the justice of the bargain comes to an abrupt end once the parties agree that, should it breach, one party shall pay a penalty to the other. In its original form, the penalty jurisdiction struck down any clause which required one party to pay to the other a sum of money upon breach which was not a genuine pre-estimate of the loss that would be suffered as a result of such a breach.③Dunlop ν New Garage [1915] AC 79.While the scope of the jurisdiction has recently been somewhat reduced, under the influence of administrative law,④Caνendish Square Holding BV ν El Makdessi, ParkingEye Ltd ν Beaνis [2015] UKSC 67, [2016] AC 1172.it is undeniable that English law dramatically restricts the parties’ freedom of contract when it comes to their secondary obligations, i.e., the obligations that obtain once the contract has been breached rather than performed. It is intriguing to note that most civilian legal systems do not have similar rules. One explanation for this may be that they tend to rely on orders for specific performance as the primary remedy for breach of contract, while English courts, in the exercise of their equitable jurisdiction, will only order one party to perform where damages are an inadequate remedy, and even then only in carefully circumscribed circumstances. This remedial regime would be undermined if the parties were allowed to agree that, in the event of non-performance, extortionate penalties were to be payable. However, a more convincing explanation is simply that most civilian systems allow courts to regulate the parties’ bargain to a greater or lesser extent, setting aside extortionate bargains or even rewriting the parties’ agreement to make it fairer. As this can be done with all the terms of the contract, there is no need for a special rule when it comes to secondary obligations. Again, a pattern emerges of a handsoff, adversarial attitude to contracting that hits a hard wall where freedom of contract is perceived to go too far.

There are some situations in which the adversarial approach of English law is simply not appropriate. These include cases in which the parties are clearly of unequal bargaining strength—be it because one of them is a consumer, be it because one of them is inherently disadvantaged by being put under duress or by being in a relationship with the other that English law characterizes as a “relationship of influence.” But even where the parties can negotiate at arm’s length, the nature of the relationship or of the tasks that one of them is supposed to undertake on behalf of the other, calls for a different, more collaborative approach. The obvious example is the contract of employment. Clearly, while prospective employers and employees may drive a hard bargain until the contract is formed, once it is, they both owe duties to one another that are absent from ordinary contractual relationships. It is thus very well established that an employee owes a duty “to serve his employer loyally and not to act contrary to his employer’s interests.”①M(fèi)alik ν Bank of Credit and Commerce International SA [1998] AC 20.This is because the contract of employment cannot reasonably be performed in any other way. Once the employee has decided to accept employment, he is on the “employer’s team” and is expected no longer to behave in an adversarial way vis-à-vis the employer.

Some employees, of course, are agents of their employers, but this is not true of all employees. Conversely, not all agents are employees: They may be acting under an independent contract or under no contract at all. Before we can usefully discuss the role that fiduciary obligations play in the law of agency, it is necessary to give a short overview of the English law of agency. This will be done in the next section of this paper.

Overview of the English Law of Agency

The institution of agency is fundamental to any advanced economy. It mainly allows commercial people to delegate the task of entering into contracts at scale or at a distance to others. At the heart of agency lies the concept of authority. Authority allows the agent to affect the legal position of the principal, most commonly by entering into a contract with a third party. Before we turn to the “agency triangle”," that is, the three distinct relationships arising in an agency situation, a few words must be devoted to peculiarities of English law, namely, the surprising lack of formality requirements for the appointment of agents and the, even more surprising, lack of a requirement that the agent be openly acting on behalf of a principal, the so-called doctrine of the undisclosed principal.

Formalities

There is no general formality requirement when it comes to the appointment of agents: They can be appointed orally (even impliedly), in writing, or in a deed (when their authority will normally be known as a “power of attorney.”) In contrast to many civilian legal orders, English law does not require the appointment of the agent to follow the same formalities as those that are required for the transaction that the agent is meant to enter into on behalf of the principal. Thus, while s. 2 of the Law of Property (Miscellaneous Provisions) Act 1989 requires that contracts for the disposition of land be in writing and signed, it allows for the signature to be executed by agents that is not required to be authorized in writing.②McLaughlin ν Duffill [2008] EWCA Civ 1627; [2010] Ch.1. In practice, however, where a lease for more than three years is granted otherwise than by deed, it will nevertheless take effect in equity: Walsh ν Lonsdale (1882)21 Ch D 9.There is one notable exception to this general rule: Only agents authorized by deed are able to execute a deed on behalf of the principal.①Powers of Attorney Act 1971 ss.1 and 7.The deed is the strongest formality requirement known to English law, yet to civilian lawyers, its requirements are surprisingly weak: In the modern law, all that is required for a deed to be validly executed is that it says that it is executed as a deed, signed and witnessed by one witness, and delivered to the donee. Deeds are only rarely required in English law for a transaction to be binding; they are useful for making binding gift promises; that is,promises unsupported by consideration, and, technically, are required to grant leases for more than three years under ss. 52 and 54(2) Law of Property Act 1925.②McLaughlin ν Duffill [2008] EWCA Civ 1627; [2010] Ch.1. In practice, however, where a lease for more than three years is granted otherwise than by deed, it will nevertheless take effect in equity: Walsh ν Lonsdale (1882)21 Ch D 9.

Where authority is given in the form of a deed, this will normally take effect as a“power of attorney.” What a “power of attorney” actually means is not defined in English law, but it can be taken to refer to the grant of a formal power, especially where that power is very wide. The Powers of Attorney Act 1971 s.1(1), as amended by the Law of Property(Miscellaneous Provisions) Act 1989, requires that instruments creating powers of attorney be executed as deeds. The authority thus conferred may be exhaustive, entitling the attorney to do everything the principal can do himself (the only exception appears to be the contraction of marriage), or it may be limited to certain defined objects. For the most part, however, it is not necessary for an agent to be invested with a power of attorney in order to have a perfectly good and valid authority so that any acts within that authority are binding on the principal. In practice, however, the power of attorney provides the agent/attorney with a document defining the extent of his authority, which he can produce as evidence to the third parties with whom he is to deal.③There are a number of specialized powers of attorney which afford the attorney more far-reaching powers than those possessed by an ordinary agent: first, a trustee (himself a fiduciary) is generally barred from sub-delegating his powers but is nevertheless able to grant powers of attorney in certain circumstances;secondly, the Mental Health Capacity Act 2005 introduced the “l(fā)asting power of attorney” (replacing the so-called “enduring” power of attorney available between 1986 and 2005). This power survives the principal’s incapacity (in contrast to an ordinary agent’s authority, which lapses when the principal, for example, becomes of unsound mind), and enables natural persons to entrust their affairs to an attorney of their choice (as long as they still have the capacity to do so), also extending to decisions about their welfare and medical treatment.

The Undisclosed Principal Doctrine

The doctrine of the undisclosed principal allows an agent to enter into contracts with third parties ostensibly on his own behalf, however, in reality, on behalf of his undisclosed principal. It is well established that the principal will be able to sue the third party directly on such contracts and vice versa. This result has been called “surprising”(Watts & Reynolds, 2021, para 8-071). in that it does not sit at all well with the general rules of contract law: The third party never makes any promise to the supposed principal,nor does the principal ever express his intention to be bound to the third party. Many have tried and failed to explain the doctrine as part of the law of agency.④See, e.g., Randy E. Barnett, Squaring Undisclosed Agency Law with Contract Theory, 75 California Law Review, 1969 (1987) (he does not really square it at all); W. Müller-Freienfels, Comparatiνe Aspects of Undisclosed Agency, 18 Modern Law Review, 33 (1955).I would suggest that an undisclosed agency be mainly used in order to protect the principal and third party against the insolvency of an intermediary who would otherwise be directly liable on the relevant contract. The language and conceptual framework of the agency are convenient tools to bring about a reallocation of insolvency risks and very little more. That this is so follows from an examination of the practical goals of undisclosed agency, of its historical development, and its current rules.①See Thomas Krebs, Some Thoughts on Undisclosed Agency, in Louise Gullifer and Stefan Vogenauer eds. English and European Perspectiνes on Contract and Commercial Law: Essays in Honour of Hugh Beale, Hart Publishing (Oxford), at 161-182 (2014).

There are essentially two situations in which there will be a commercial need to hide the fact that the intermediary is acting on behalf of somebody else:②See Peter Watts and F. M. B. Reynolds, Bowstead & Reynolds on Agency, Sweet & Maxwell (London), para 8-073 (2021).The principal may wish to enter the market without this being known (for a variety of reasons—frequently because he places a premium on goods which would be reflected in the price were his identity known); similarly, the intermediary may wish to avoid that the third parties he is dealing with will in future deal with the principal directly, cutting him out of the transaction. These goals could be achieved by a chain of contracts (P-A-T), with each party being liable and entitled to the person next to him in the chain only, and this is the solution adopted in civilian jurisdictions under the heading “indirect representation” or“commission agency.” These legal systems then provide rules which, to a greater or lesser extent, make the third party directly liable to the principal if the intermediary becomes insolvent, so that the difference between those systems and the common law is not as great as may, at first sight, appear (K?tz, 1996, p. 367; Watts & Reynolds, 2021, p. 377).Historically, the undisclosed agency doctrine appears to have been developed to deal with precisely the insolvency of an intermediary who had been acting in his own name but on somebody else’s account (Stoljar, 1961, pp. 203-211). The modern doctrine bends over backward to protect the third party from nasty surprises, so that its rules are a long way removed from the rules applicable to disclosed agency scenarios. Thus, the undisclosed principal is barred from ratifying contracts entered into by the agent in excess of his authority—had the agency been disclosed, he would be able to do so.③See below.In addition, the agent does not “drop out”—he remains liable and entitled alongside the principal. The third party can avail himself of any defenses—including set-off—which he would have against the agent. Finally, the contract may, by its express or implied terms, exclude the possibility of an undisclosed principal. It is thus arguable that undisclosed agency has rather more in common with assignment than it does with agency proper. Indeed, when they were consulted on the doctrine by the drafters of the then-nascent Unidroit Principles of International Commercial Contracts, the two foremost experts on the common law of agency, Professors Francis Reynolds of Oxford University and Deborah Demott of Duke University, thought that nothing of great value would be lost if undisclosed agency were not included in the new international code.

Still, the practical problems which might arise in commission agency transactions where the agent becomes insolvent having been put in funds by either principal or third party might nevertheless be thought to call for the law’s intervention. If the commission agent is regarded as a mere conduit pipe between the principal and the third party, there are arguments why the parties to the transaction should not be exposed to a greater insolvency risk than if they had been dealing directly with each other: Though not acting in his principal’s name, the commission agent is acting on his principal’s account; he is not taking any risks with his own money and is, as the name implies, paid by way of commission rather than by generating profits from the transaction for himself. On the other hand, of course, the principal is using the commission agent for a reason, knowing that he may fail and choosing to use him anyway, while the third party does not even know that the commission agent is not acting on his own account. Preferring the principal and/or third party in the intermediaries’ insolvency is not a foregone conclusion. In other words, the arguments are finely balanced.

The problem arises most acutely when the commission agent enters into a contract with the third party and fails before the third party performs. The trustee in bankruptcy cannot necessarily be relied on to enforce the contract on the principal’s behalf and may, in fact,seek to retain any payments made by the third party for the benefit of the bankrupt estate.Conversely, if the third party grants the intermediary credit, while the principal may well be able to reap the benefits of the third party’s performance, the third party will be hardpressed to get his money from the principal and may, for that reason, be unwilling to deal with the intermediary.

Some jurisdictions tackle this problem by statutory rules by which the principal and the third party are brought into a direct contractual relationship in such cases.One example of this can be found in the Principles of European Contract Law (PECL).Where an intermediary has become insolvent (or it is clear that he will not perform for some other reason), the principal may exercise any rights acquired by the intermediary against the third party on his behalf (Art. 3:302). By the same token, the third party may proceed directly against the principal and exercise any rights which he has acquired against the third party (Art. 3:303). In both cases, this is, of course, subject to defenses, which the third party could have relied on. These provisions are modeled on similar rules, which may be found in a number of jurisdictions, in particular the Netherlands, Belgium, Denmark, Sweden, Italy, and Portugal.①See Ole Lando and H. G. Beale eds. Principles of European Contract Law: Parts I and II, Kluwer Law International (The Hague), at 222 (2000).A common law system, developing organically, cannot easily put in place similar rules. The common law, therefore, solved the problem by developing, very early on, a doctrine by which a contractual relationship between the undisclosed principal and the third party is brought about directly. The practical results will generally be indistinguishable from those reached in other jurisdictions.

The new Chineseciνil codedoes not subscribe to the undisclosed agency doctrine.Instead, it follows the Principles of European Contract Law in denying a direct nexus between the (undisclosed) principal and the third party unless and until something goes wrong. Thus, Art. 162 clearly expresses the “publicity principle”: The agent’s act will only bind the principal where it is carried out “in the name of the principal.” Art. 925 modifies this to some extent, in that it puts in place a presumption in favor of agency where the third party knows that the person he is dealing with is another party’s agent. While this moves the law a little bit in the direction of the undisclosed agency, it is, in reality,no more than an interpretative guide. Art. 926, on the other hand, is a novel provision which allows the principal or the third party whose only nexus is the engagement with a commission (i.e., indirect) agent to intervene in the commission agent’s contract where the other, that is, the principal or the third party, has prevented the commission agent from performing. In English law, the direct nexus would arise at a prior point through the undisclosed agency doctrine. The Chinese solution is, however, also narrower than the solution adopted by the PECL, in that it does not cater to the situation, in which the agent does not perform for reasons related solely to himself.

The “Agency Triangle”

The three relationships involved in agency form the so-called “agency triangle,” which results in a binding, bilateral contract between the third party and the principal. They are principal/agent, agent/third party, and principal/third party. We will briefly examine each in turn before then examining the principal/agent relationship in more detail, given the focus on the fiduciary nature of the relationship adopted by this paper.

Principal and Agent.

The relationship between the agent and the principal may (and usually is) but need not be contractual. The peculiarly common law doctrine of consideration, which insists on a counter-performance, or the promise thereof, to support a promise, means that it is entirely possible for an agent to act on behalf of a principal without being contractually obliged to do so. His failure to act, where he has undertaken to act without being promised anything in return, might, of course, lead to losses being incurred by the principal—however, the agent’s liability would, in such a scenario, be tortious only.

It is a fertile source of confusion to equate the contract the principal might have with the agent with the scope and extent of the agent’s authority. The two are analytically and logically distinct. The principal, in asking the agent to act, will tell him, expressly or impliedly, what the extent of his authority is to be. This is entirely independent of the contract, if any, between the principal and the agent. Should the agent then exceed his authority in a way that exposes the principal to liability (on which more below), that may be a breach of contract, if a contract between principal and agent exists, or expose the agent to liability in tort (for negligence, a duty of care arising from his voluntary assumption of responsibility, on which more below).

The agent will also be expected, of course, to execute his mandate, in other words,to do what he undertook to do vis-à-vis the principal. Where he is in a contractual relationship with the principal, the default position is that he will only be liable if his failure to achieve the desired outcome is due to his failure to carry out his mandate “with reasonable care and skill,”①Supply of Goods and Services Act 1982, s. 13.although it is, of course, entirely possible for an agent to agree to bring about a certain result, failing which he incurs liability.

In English law, it is seen as fundamental to the relationship between principal and agent that the agent acts in a “fiduciary” capacity, subjecting the agent to sweeping and not necessarily well-defined duties accompanied by draconian remedies. We will look at this in detail below. For now, we should note that such a broad label is conspicuously missing from the ChineseCiνil Code. It is also remarkable that the agent is made to such sweeping and unforgiving duties within the context of a legal system that normally avoids the use of wide principles of this sort, most obviously, of course, the refusal of the English law of contract to recognize a general principle of good faith.

The justification for making the agent a fiduciary is that, by granting authority to bind him in transactions with others, the principal is putting himself, to an extent (to be precise, to the extent of the authority granted to the agent or represented to third parties)at the mercy of the agent. The adversarial model explored above is no longer appropriate,and the pendulum swings the other way dramatically. Agent and principal are on the same team, and as such, the agent is not supposed to be acting in his own interests and against the interests of the principal, but exclusively in the interests of the principal. All temptation to do otherwise is removed by making the agent accountable with respect to all profits, over and above commission earned from the principal, that the agent may have derived from his agency in breach of his fiduciary duties.

Agents are not the only fiduciaries in English law. In fact, the paradigm fiduciary is the trustee. The institution of the trust is generally regarded as the crowning achievement of the Court of Chancery and the legal system that, for centuries, co-existed with the common law: equity. Put at its simplest, one person (the settlor) transfers property to another (the trustee) who agrees to hold it on behalf of a third (the beneficiary). While civilian systems, with greater or lesser success, try to achieve a similar result by using the law of obligations, maybe supplemented by tweaking the rules applying should the trustee become insolvent, English law treats the beneficiary as the owner, in equity, of the trust property, while the trustee has the “l(fā)egal” title—he is the owner “at common law,” but can be ordered by courts of equity to transfer the property to the beneficiary in appropriate circumstances. While the two systems co-existed separately until 1877,they have, since the Judicature Acts 1877, been administered by the same courts, and this adds a complication to the study of English law that many foreign exchange students are unprepared for. The reason why the trust is mentioned here is that it raises similar concerns to agency: the trustee is endowed with the legal title, a title that, should he decide to abuse the “trust” placed in him, he could validly pass to others (so long as they are unaware of the trust). Like an agent, the trustee is therefore required to subordinate his own interests to those of the beneficiary—he is subject to fiduciary duties. These duties were developed by courts of equity over centuries and could easily be transplanted into the agency relationship. In fact, an agent is often treated exactly as if he were a trustee.

Legislative intervention by the European Union and with a long history in civilian legal systems has recently introduced rules into English law that recognize that while the principal may, to some extent, be at the mercy of the agent, the converse is also the case. Say an agent is employed to seek out customers for the principal, using his existing relationships, his knowledge of a given market or his peculiar skills, to build up a customer base in a given market. Once this has been achieved, the principal dismisses him and takes over the customer base directly, cutting out the “middle man,” as the cliché goes. English law traditionally did not see this as a problem—the parties were dealing at arm’s length, and if the agent did not protect his position by the terms of the contract,he only had himself to blame. Civilian legal systems saw this differently, recognizing parallels between what they termed “commercial agency” and employment relationships,and since the European Union took a dim view of divergent rules in different member states, it sought to harmonize the position in Directive 86/653/EEC (coordination of the laws relating to self-employed commercial agents), and this was implemented in England by the Commercial Agents (Council Directive) Regulations 1993. The fate of these Regulations hangs in the balance following the United Kingdom’s withdrawal from the EU, but for now, they remain part of English law and require a principal who wishes to terminate his relationship with a “commercial” agent to “indemnify” or “compensate”him. Unlike the agent’s fiduciary duties, these cannot be contracted out.①M(fèi)orland J explained the policy behind the Directive and the Regulations implementing it as follows: “It might take years for an agency to be developed to a state of profitability. If then terminated the commercial agent loses his livelihood. The resources and effort put into the development of a profitable agency are lost to him whereas the principal gains a valuable asset.” (Ingmar GB Ltd ν Eaton Leonard Inc [2001] CLC 1825 [33]. The Directive left it to the Member States whether to focus on the agent’s loss or the principal’s (unjust) enrichment, but, to the chagrin of legal advisers, the UK Parliament left both options open. Anecdotally, the most frequent question I am asked when giving talks to practitioners on agency is how the Regulations can be avoided!

Agent and Third Party.

In most cases, the agent will not end up in a legal relationship with the third party: He“drops out” of the picture. There are two exceptions to this. The first involves situations in which the agent exceeds his authority, with the result that the third party does not,in fact, end up in a contractual relationship with the principal. The agent may, in such a scenario, be liable to the third party for breach of his “warranty of authority,” the precise scope and legal nature of which is still subject to some doubt in English law.②It could be explained as a claim on a genuine “warranty,” i.e., a collateral contract by which the agent agrees to indemnify the third party against the consequences of any lack of authority on his part, or it could be seen as a tort claim based on the agent’s misrepresentation to possess authority he does not,in fact, possess. The problem with the former is that imputing an intention to create legal relations (between himself and the third party) may be difficult,while the problem with the latter is that normally English law does not hold parties strictly liable for the veracity of their statements, requiring some lack of care on the part of the agent before holding him liable, while breach of warranty of authority does not depend on fault: Collen ν Wright (1857) 7 E & B 301.An additional problem with the tort analysis is that it is well established that the third party’s expectation interest is protected. See further F. M. B. Reynolds,Breach of Warranty of Authority in Modern Times, 1 Lloyd’s Maritime and Commercial Law Quarterly 189 [2012].The second is more straightforward: The agent agrees to be liable to the third party and is entitled to enforce the contract alongside the principal. If neither of these applies, the agent indeed“drops out” as far as the third party is concerned.

Principal and Third Party.

The whole point of the agency is to bring the principal into a contractual relationship with a third party. The legal mechanism by which this occurs is referred to as authority.An agent’s authority gives him the power to bind the principal. So how is the agent endowed with the principal’s authority, and why does authority matter?

There are a number of different ways, in which the different ways, in which an agent can be clothed with authority, can be categorized. In English law, these tend to cut across one another: We speak of actual and apparent, express, implied, and usual authority. This can be quite confusing. Leaving usual authority to one side (it overlaps with all four other kinds of authority), we shall focus on actual and apparent authority, both of which can be founded expressly and impliedly. Finally, even where the agent is entirely unauthorized,the principal may have the power to cure this lack of authority by ratifying the agent’s unauthorized act (as is also the case in Chinese law, of course).

Actual Authority.The term “actual” authority is misleading. It implies that other forms of authority are not “actual,” and are not “actually” effective. And yet it is undeniable that a principal will end up liable to a third party who contracts with an agent clothed with merely “apparent” authority. What the law means by the label “actual” is,therefore, something else. Where the agent has actual authority, his power to affect his principal coincides with his right to do so. Both power and right must be gleaned from what the principal said to and did vis-à-vis the agent, and this can only be done by a process of interpretation and construction. This process may be very straightforward and clear, as when the principal tells the agent to buy 100mt of a certain grade of grain on the commodities market at a certain price. But even then, of course, the language used by the principal is being interpreted, and the process of interpretation is easy. In such cases, we refer to “express actual authority.” But as the language used is less clear,less certain, where, perhaps, the agent is given a task (such as “get me a good deal on the grain”), much more work is done by the process of interpretation, and aspects of the agent’s authority will be implied. It is in such cases that we talk in terms of “implied actual authority.” The process of construction is the same as in the general law, with one important and easily overlooked difference: As the principal’s interlocutor is the agent,and not directly the third party with whom he will end up in contractual relations, the process of interpretation is carried on by the agent and from the agent’s point of view,with his, the agent’s knowledge of the surrounding facts and on the basis of any course of dealing principal and agent, and not principal and the third party, might have had.①Cf. to the same effect, Howard Bennet, Principles of the Law of Agency, Hart Publishing (Oxford), at 40 (2013).

One more point needs to be made under this heading. The agent knows that he is employed to act on his principal’s behalf and in the principal’s best interests. Any act that the agent undertakes that the agent knows to be not in the principal’s best interests is,by definition, not covered by the agent’s actual authority. This principle was set out by O’Connor J inLysaght & Co Ltd ν Falk:②Lysaght & Co Ltd ν Falk (1905) 2 CLR 421, 439.“Every authority conferred upon an agent, whether express or implied, must be taken to be subject to a condition that the authority is to be exercised honestly and on behalf of the principal. That is a condition precedent to the right of exercising it, and, if that condition is not fulfilled, then there is no authority, and any act purporting to have been done under it unless in dealing with innocent parties, is void.” The qualification “unless dealing with innocent parties,” may give us pause for thought. It is generally thought③Cf. Peter Watts and F. M. B. Reynolds, Bowstead & Reynolds on Agency, Sweet & Maxwell (London), para 3-012 (2021), where the above passage is cited and commented on. This is where the label “usual authority” is sometimes encountered, although, it is suggested, the label should be avoided as it is inherently imprecise: it can refer to both apparent and implied actual authority, depending on the circumstances.that this refers to the agent’s apparent authority, and we shall return to it in that context. An example of the principle can be seen in the Australian case ofSweeney ν Howard:④Sweeney ν Howard [2007] NSWSC 852. American Law Institute, Restatement of the Law Third, Agency, American Law Institute (St. Paul), § 2.03, Comment (c) (2006).there, a power of attorney conferred very wide-ranging authority on the agent. The agent used it to borrow money secured by a mortgage on his principal’s land, intending all along to lend it on for the agent’s purposes. This was held not to be covered by the terms of his actual authority: The agent knew he had no business exercising his authority for his own ends.①See also Hopkins ν TL Dallas Group Ltd [2004] EWHC 1379; [2005] 1 BCLC 543 [89].In such cases, the focus shifts to the agent’s “apparent” authority.

Apparent Authority.Where the agent’s power to bind the principal is not matched by a corresponding right to do so, the law speaks of “apparent” authority. This is analogous to Art 172 of theCiνil Code,which provides that an “act of agency” will nevertheless be valid even where the agent has no “power of agency.” “Power” seems to refer to the concept of authority because if the act is valid, it seems clear that the agent did, in fact,have the necessary “power”—he just was not, as between himself and his principal,permitted to use it. The validity of the (unauthorized) act depends on the third party having “a good reason to believe that the actor has the power of agency.” The protection afforded to the third party, however innocent, seems excessive. English law is rather more conservative. It is well-established that there must be a representation by the principal for the principal to be exposed to liability for the agent’s unauthorized acts.②Freeman & Lockyer ν Buckhurst Park Properties (Mangal) Ltd [1964] 2 QB 480.This may, of course, consist of the simple act of putting the agent into a certain position, a position, that is, that would normally be assumed to be accompanied by a certain level of authority.③Cf. Peter Watts and F. M. B. Reynolds, Bowstead & Reynolds on Agency, Sweet & Maxwell (London), para 3-012 (2021), where the above passage is cited and commented on. This is where the label “usual authority” is sometimes encountered, although, it is suggested, the label should be avoided as it is inherently imprecise: it can refer to both apparent and implied actual authority, depending on the circumstances.

English law insists on representation because it regards the basis of apparent authority to be the doctrine of estoppel. An estoppel is basically a rule of evidence that prevents a party from leading evidence that contradicts a statement he himself made to the other party that was then relied on by that party: He cannot show, by evidence, that he was, on that occasion, a liar. The American Restatement of Agency eschews this complexity,④Sweeney ν Howard [2007] NSWSC 852. American Law Institute, Restatement of the Law Third, Agency, American Law Institute (St. Paul), § 2.03, Comment (c) (2006).and rightly so, as I have argued elsewhere: in the context of a law of contract that looks to objective manifestations of consent, it is not necessary to use the artifice of estoppel(Krebs, 2010, pp. 205-224). Yet English cases seem committed to this idea,⑤Rama Corp Ltd ν Proνed Tin & General Inνestments Ltd [1952] 2 QB 147, 149; Freeman & Lockyer ν Buckhurst Park Properties (Mangal) Ltd [1964] 2 QB 480, 504-505. It is difficult, however, to find a case in which the estoppel analysis is necessary to justify the decision, leaving an objective interpretation within interpretative reach.with the consequence that, in contrast to Art 172 of theCiνil Code, any act by the agent that is covered by his apparent authority is binding on the principal only—without more, the principal does not derive any rights against the third party from the agent’s act. However,the practical effect of this should not be overstated, given that the principal will normally be able to ratify the agent’s unauthorized act, thus bringing about a bilateral contract between himself and the third party.

Both American and English law do, however, require a contribution of the principal that gives rise to the third party’s belief that the agent is authorized, and here they are both narrower than Chinese law appears to be. Again, however, the representation in question can be express or implied. While it is unusual to speak of “express apparent authority” or“implied apparent authority,” such a distinction would be useful to distinguish between direct communications of the scope or breadth of the agent’s authority (which are,however, contradicted by internal restrictions placed on the agent’s actual authority) and implied statements made, for example, by placing the agent in a position in which it would be usual for the agent to possess the sought-for authority.

This becomes clear in the so-called “self-authorization” cases. The first in this line of cases is The Ocean Frost,①Armagas Ltd ν Mundogas SA [1986] AC 717.another shipping case. Both contracting parties were represented by utterly corrupt agents. They both knew that the one acting for the buyer of a ship had no authority whatsoever to commit his principal to a three-year, rather than a one-year, time charter of the ship. The House of Lords acknowledged in the course of the judgment that it was possible for an agent to be authorized to communicate the extent of authority possessed by another agent and averted to “rare and unusual” cases in which an agent, lacking authority to enter into a certain transaction, might be authorized to communicate the principal’s representation that he had been authorized to enter into that self-same transaction. This was not so in the case itself, and the argument that the corrupt agent had been able to “self-authorize” in this manner was rejected. However,some seven years later, the Court of Appeal came to decide the case ofFirst Energy (UK)Ltd ν Hungarian International Bank Ltd②First Energy (UK) Ltd ν Hungarian International Bank Ltd [1993] 2 Lloyd’s Rep 194.on the basis of the “rare” principle set out by the House of Lords in The Ocean Frost. A Hungarian merchant bank had a head office in London and a further office in Manchester. The manager of the Manchester office had dealt with the third party throughout the transaction on behalf of the bank, and the third party knew that it went beyond his authority to grant loans above a certain amount. When he approved a loan on behalf of the bank, assuring the third party that he had received the requisite one-off authority to do so, the Court of Appeal held that he had been able to make a representation that gave rise to his own apparent authority. The crucial aspect of the case is that, normally, a branch manager of a bank would have the authority to approve loans,and that the third party only knew about his lack of authority because he had told them.Having taken away his own apparent authority, he was in a position to reinstate it. The case is nonetheless not uncontroversial and has not been followed in Canada③British Bank of the Middle East ν Sun Life Assurance Co of Canada (UK) Ltd [1983] 2 Lloyd’s Rep 9.and Singapore.④Skandinaνiska Enskilda Banken AB ν Asia Pacific Breweries (Singapore) Pte Ltd [2009] SGHC 197.

This is as far as English law is prepared to go when it comes to “self-authorization.” It is not enough for the agent to claim to be authorized, however respectable or trustworthy he may appear to be. English law takes a stance that is rather less forgiving than the ChineseCiνil Codein Art. 172, unless that provision is interpreted restrictively so as to require a causal link of some sort between the principal and the “good reason” to believe the agent’s authority on the part of the third party.

There are some cases where it is doubtful if an English court would find the requirements of apparent authority to be made out. Say an agent is issued with visiting cards by his principal, showing him as a “senior buyer” of the principal. Clearly, any transaction that would normally be within the ambit of a “senior buyer’s” authority will be covered by his apparent authority, even if he exceeds the scope of his “actual” authority,the requisite representation by the principal being present in the form of the visiting card. Would and should it make a difference, then, if the agent ordered the cards himself,as long as his title of “senior buyer” was correct? A case that might help in that sort of situation, but which has long been doubted in the literature, isHambro ν Burnand.①Hambro ν Burnand [1904] 2 KB 10.There, an agent, writing insurance, exceeded his authority by acting in his own, not his principal’s best interests. His authority, however, had been reduced to writing, and he would have been able to present this to the third party, thus clothing himself in apparent authority. The court did not draw a clear distinction between actual and apparent authority in holding the principal bound to the third party. The reason for this might well have been a reluctance to insist on a direct representation that would normally be required for apparent authority in circumstances where an appearance of authority could have been created by asking the agent to show his credentials. Again, a court applying Art 172 would probably reach a similar result; however, much of this is criticized in English law.

Ratification.Ratification allows a principal to adopt the transaction of an unauthorized agent. He can thus take the benefits of that transaction, clothing the agent with authority retrospectively. In practice, the need for ratification may arise in a variety of circumstances. The agent may not realize that he is exceeding his authority, or that he has no authority at all.②It is the object of the rule in Ireland ν Liνingston (1871-72) LR 5 HL 395 to prevent this, but the rule is said to be of limited applicability in modern times, given the increased technological scope for referring back to the principal: see e.g. European Asian Bank AG ν Punjab and Sind Bank (No 2) [1983] 1 WLR 642, 655.It may also be that, though aware of his lack of authority,the agent believes that the transaction is in his principal’s best interests, so the principal would have conferred the requisite authority on the agent had he been aware of the circumstances. Finally, the agent may enter into the transaction knowing full well that it is not covered by his authority, and that it would be against the principal’s likely wishes.Particularly in the first two situations, the usefulness of the doctrine of ratification is apparent: It enables the principal to take advantage of the agent’s transaction without needing to enter into a new contract; if the third party seeks to take advantage of the agent’s alleged lack of authority in order to escape from a bad bargain, the principal may seek to put the matter beyond doubt by ratifying the transaction.

One important limit of ratification is that the agent must have been acting on behalf of another—in other words, the undisclosed agency is excluded. The reason given for this rule by the House of Lords in the leading case,Keighley, Maxsted & Co ν Durant,③Keighley, Maxsted & Co ν Durant [1901] AC 240.is that“civil obligations are not to be created by, or founded upon, undisclosed intentions.”④Keighley, Maxsted & Co ν Durant [1901] AC 240, at 247 (Lord Macnaghten).The editors of Bowstead & Reynolds on Agency criticize this reasoning, and the unavailability of ratification, as inconsistent with the very idea of the undisclosed agency, which has as its foundations precisely the agent’s undisclosed intentions (Watts & Reynolds, 2021,para 2-061). I am going to put forward two counter-arguments, one on a practical and one on a conceptual level. The first argument, which was clearly strongly influential on the House of Lords inKeighley ν Durant, is the danger of ex post fabrication. Granted, this danger exists in the context of undisclosed agency as well, and the whole doctrine of the undisclosed principal can be criticized on this ground. However, where an undisclosed agent acts with prior authority, any evidence that such authority was actually granted will clearly be more reliable than in the case of subsequent ratification. It will involve some sort of communication between the principal and the agent, possibly even in writing. On the other hand, the adoption of the contract by the principal in cases where the agent lacked authority at the time of the contract will be possible as long as the agent can credibly assert that, when entering into the contract, he intended to do so on the principal’s account. This seems a much lesser burden than that of establishing antecedent authority,so that the danger of fabrication is correspondingly greater, with the result that the law would, in the words of Lord Shand, “give one of two contracting parties in his option,merely from what was passing in his own mind and not disclosed, the power of saying the contract was his alone or a contract in which others were bound with him.”①Keighley, Maxsted & Co ν Durant [1901] AC 240, at 250.

Reynolds finds it surprising that the rule applies even where it is sought to hold the ratifying principal liable, in other words, where obligations are imposed rather than rights conferred on him.②Hugh Beale ed. Chitty on Contracts (33rd edition), Sweet & Maxwell (London), vol. 2, at 17, fn. 151 (2020).Keighley ν Durantwas itself such a case: The seller of wheat unsuccessfully sought to recover damages for non-acceptance from the alleged principal,who had privately agreed with the agent to take the contract on a joint account a day after it had been concluded in the agent’s name alone, and in excess of the price at which he was authorized to buy on the joint account. The Restatement 3d relies on a number of US cases which appear to contradict this result, but none of these appear to be a particularly strong authority to support a departure from a rule well-established by English authority and accepted by the first two Restatements of Agency.③Coyle ν. Smith, 300 So. 2d 738 (Fla.Dist.App.1974); Young & Rubicam, Inc. ν. Ticket Holder Marketing, Inc 1989 WL 4210 (N.D.Ill.) and Acuri ν. Figliolli, 91 Misc.2d 831, 398 N.Y.S.2d 923 (N.Y.Dist.Ct. 1977).Two out of three of these cases involve alleged acts of ratification by undisclosed principals who accepted and retained money originating from the third party, which was paid into their bank accounts.④In Young & Rubicam ν Ticket Holder, the proceeds of a bill of exchange drawn on the third party were paid into the alleged principals’ bank account. In Acuri ν Figliolli, the alleged undisclosed agent had sold his mother’s car, accepted a number of installments in payment, which he had paid over to the mother, and later repossessed the car for an alleged breach of contract. The third party’s claim against the mother was for restitution of the instalments paid, and the action was, in fact, an action for money that had and received.The cases cited in the Restatement are, to some extent, consistent with the proposition that an undisclosed principal may well be bound by an act amounting to ratification where his liability arises in tort or unjust enrichment—neither cause of action depends on showing a consensus ad idem of any kind between the principal and the third party. Clearly, where the principal asserts the rights of an owner over property belonging to the third party or retains a benefit at the third party’s expense in circumstances which the law considers unjust, it is right and proper that he should be held liable accordingly, be he undisclosed or disclosed. This does not, however, change the general rule that neither he nor the third party will be bound in contract.

Ratification is said to take effect retrospectively—in other words, the law pretends that the agent had authority all along. The practical effect of this is that even when the third partydiscovered the agent's lack of authority and made it clear that he no longer wishes to be bound, the principal is nevertheless free to ratify and bind the third party to the contract. If this is correct, it is difficult to square ratification with ordinary contract doctrine: At no point are the parties “ad idem”; at first, the principal has not expressed his intention to be bound—later on, the third party has expressly disavowed any such intention. The leading English case which is usually cited as authority for the above proposition isBolton ν Lambert.①Bolton ν Lambert (1889) 41 Ch D 295.It concerned a contract to take a lease. Lambert wrote to Scratchley, a director of Bolton Ltd, offering to take a lease from the company. Scratchley replied to Lambert, accepting this offer. The very next day, Lambert sought to withdraw from the contract. It appears from the report that he did not at that point know that, in accepting the offer, Scratchley had exceeded his authority. When Bolton Ltd later sued for specific performance, this circumstance became clear, and Lambert argued that he had,on that ground alone, been entitled to withdraw from the contract. The Court of Appeal disagreed and held that the contract was valid. It, therefore, seems that the rule inBolton ν Lambert, to the effect that ratification of a contractual offer or acceptance is retrospective,is binding on English courts up to the Court of Appeal. Cotton LJ appreciated the practical difficulties which the fiction that ratification had retrospective effect might cause, and “how favorable the rule was to the principal, because till ratification he was not bound, and he had an option to adopt or not to adopt what had been done.”②Bolton ν Lambert (1889) 41 Ch D 295, 307, citing Hagedorn ν Oliνerson (1814) 2 M & S 485, an insurance case in which Lord Ellenborough CJ pointed out that the retrospective effect of ratification meant that a principal could avoid paying an insurance premium where no loss was suffered, while taking the benefit of the insurance in the event of a loss by ratifying his agent’s unauthorized entering into the policy on his behalf. However, that situation did not, in fact, arise in that case, and as such, the dicta relied on by Cotton LJ are obiter.He then went on to assert, however, that he was bound to apply the fiction and find in favor of the principal.

The problems with the fiction inBolton ν Lambertflow from the fact, recognized by Isaacs J inDaνison ν Vickery’s Motors Ltd,③(1925) 37 CLR 1.that the case is simply irreconcilable with the ordinary rules of contract law. Assume R, a rogue, writes to S, offering to buy goods at a certain price and forging the signature of C, a long-standing customer of the company. S writes to C, accepting the offer. It is now, beyond any doubt, following Shogun Finance,④Shogun Finance ν Hudson [2004] 1 AC 919.that there is no contract of any kind in this situation. However, C may well decide to treat S’s acceptance as an offer to sell it (C) the goods on the terms set out therein. It is equally clear, however, that S can withdraw that offer at any time until C has actually accepted it.Can it make any difference to this scenario if R, rather than pretending to be C, pretended to be acting on behalf and with the authority of C? The law, it is suggested, cannot possibly reach contradictory results in the two cases.

Modern Chinese law, along with most civilian codifications, thePrinciples of European Contract Law, and thePrinciples of International Commercial Contracts,avoids these difficulties in Art. 171. That provision attempts to maintain a balance between the interests of the principal and the third party by giving the third party a means by which ratification (or refusal to ratify) can be expedited or even forced: The third party can, once he finds out that the agent lacked authority, give notice to the principal asking him to clarify the position within 30 days. More importantly, the danger of creating a unilaterally-binding contract is avoided by allowing the third party to escape from the bargain, pending ratification. It is suggested that these are much more sensible rules than those of English law, but that they are rules that are not easily implemented in common law, case-based system.

Liabilities of the Agent to the Principal

As already mentioned, English law is still, at least nominally, committed to the doctrine of consideration which, in its truest form, only imposes contractual obligations on a party where the party receives some sort of counter-promise or counter-performance.This used to go hand-in-hand with the notion that there was only very limited scope for imposing private law liabilities outside of contract. This initially hampered the development of the law of tort: In the famous case ofDonoghue ν Steνenson,①M(fèi)’Alister (or Donoghue) ν Steνenson [1932] AC 562.well known even outside the common law world, the argument that a manufacturer should not be liable in tort to the ultimate consumer of a defective product (in the actual case, a bottle of ginger beer containing the decomposed remains of a snail) because she had not entered into a contract with the manufacturer and should therefore have no claim against it,almost succeeded in the House of Lords: The appeal was allowed by a bare majority, with Lords Buckmaster and Tomlin dissenting. From that case, it was still a very long road for the law to reach the position where a person can be liable not just for actions but for statements, with the decision inDerry ν Peek②Derry ν Peek (1889) L.R. 14 App Cas 337.standing as authority for the proposition that “in the absence of the contract, an action for negligence cannot be maintained where there is no fraud.”③Le Liνre ν Gould [1893] QB 491 at 498 (Lord Esher MR).In the seminal case ofHedley Byrne ν Heller,④Hedley Byrne & Co Ltd ν Heller and Partners Ltd [1964] AC 465.decided by the House of Lords in 1964, 32 years afterDonoghue ν Steνenson, it was, however, put beyond doubt that duty could arise where it was voluntarily assumed. As Professor Edelman (now Edelman J, High Court of Australia) explained in a very important article (to which we will have to refer again) in 2010, “‘contract’ does not exhaust the category of promises or undertaking having legal effect. Where the issue involves a failure to take care or skill,the only question is whether there has been a voluntary assumption of responsibility to the defendant in the performance or undertaking of a task” (Edelman, 2010, p. 302). In the context of agency, this translates into a position where a person who agrees to act as an intermediary gratuitously will be liable in tort, on the basis of having voluntarily assumed responsibility, if he performs the agreed tasks negligently or not at all, but only to the extent that this has made the principal’s position worse. Where the principal can prove, of course, that in the absence of the would-be agent’s undertaking, he would have performed the relevant task himself or found someone else to perform it, this will not look very different from the contract measure of damages—though it will, in fact, just be the tort measure,①Cf. East ν Maurer [1991] 1 WLR 461.putting the claimant in the position he would have been in had the tort (i.e., the voluntary undertaking coupled with a failure to honor it) not been committed.

Where the agent is subject to contractual duties, it is, of course, not necessary to prove reliance on the part of the principal in this way. In the vast majority of cases, the agent will be under a duty to perform his mandate with reasonable care and skill because he contractually undertook to do so.

The orthodox view is that, just by virtue of being an agent, and whether or not his relationship with the principal is contractual, a person becomes the subject of very wide and strict fiduciary duties. The most obvious difference between theCiνil Codeand English law lies in the Code’s refusal to subject agents to such very wide duties. Instead,specific duties and liabilities of the agent are provided for: Art. 164 makes an agent liable for “damage caused to the principal due to failure to perform or fully perform duties of agency,” and provides for joint and several liability of the agent and the third party where the agent “maliciously conspires with the counter-party to infringe the principal’s legal rights and interests.” Art. 168 bans self-dealing and, indeed, cross-dealing in situations where an agent represents several principals. The Code is silent on the remedial framework that will apply should those duties be breached.

As we have already seen, English law puts the agent under a more general “fiduciary duty.” We shall examine two questions in this context:(a) How does English law decide on whom fiduciary duties are to be imposed?(b) What is the content of fiduciary duties imposed on agents?

The Imposition of Fiduciary Duties

The orthodox view in English law appears to be that the imposition of fiduciary duties depends on the characterization of the relevant relationship and the status of the supposed fiduciary. This may not present many problems when it comes to the paradigm fiduciary, the trustee,②At least not where the trust is express. It has long been used as an argument against the imposition of a resulting or constructive trust that thereby the “trustee”might become subject to fiduciary duties he could have known nothing of.the picture when it comes to the agency is more complex. I have argued elsewhere (Krebs, 2010, pp. 205-224) that little hinges on the characterization of a person as an agent: While the vocabulary of authority provides useful tools in describing the incidents of the agency triangle, a contract entered into by an “agent” obeys the ordinary rules of offer and acceptance. However, if one consequence of being an “agent”is the imposition of fiduciary duties, that picture changes dramatically. Suddenly, the question of whether someone is an agent or some other form of intermediary, becomes allimportant. The problem with this is that English law does not really have a definition of“agent.” Unlike civilian systems, it does not draw a clear distinction between a messenger,an agent, and intermediaries in between these two. If the imposition of fiduciary duties is status-based, the label “agent” would suddenly assume exaggerated importance. English law has not developed a clear definition of what makes an “agent.” For example, a cashier in a supermarket is clearly able to affect the supermarket’s legal position: By ringing items up in the cash register, he will form a contract between the supermarket and the customer. And yet he will have little or no discretion, e.g., in setting the price. And yet it is that discretion, that ability to “sell the principal down the river,” that is seen as the main reason for imposing fiduciary duties.

This is one reason why Professor Edelman, in his aforementioned article (Edelman,2010, p. 302), argues forcefully and convincingly that fiduciary duties arise in the same way as other contractual (and tortious) duties: by implication, depending on the particular circumstances of the relationship. This would mean that, rather than being made subject to the whole barrage of fiduciary duties, only such duties could be implied as were appropriate on the facts.

The locus classicus for fiduciary duties in a commercial context, as Professor Edelman explains (Edelman, 2010, pp. 302, 310), isBoardman ν Phipps.①Boardman ν Phipps [1967] 2 AC 46.A solicitor to a trust and one of its beneficiaries noticed that the trust held a minority shareholding in a company that was being run in a way that they found sub-optimal. They thought that the company should be taken over, but the trust could not afford to bid for a majority stake. The solicitor and beneficiary joined forces and, using their own funds, took over the company.They turned the company’s business around and made it much more profitable, which also benefited the trust. Nevertheless, they were held liable to account for their profits, having breached their fiduciary duties to the trust by not obtaining its fully informed consent.Why did the solicitor and the beneficiary become subject to fiduciary duties? All the courts deciding the case focused on the supposed agency relationship between the trust and the defendants, even though it was fairly clear that they had not been appointed as agents to represent the trust. True, they had purported to represent the trust in shareholder meetings, but, as we have seen above, a person cannot appoint himself as an agent.Professor Edelman argues, rightly, that the reason why the fiduciary duties were imposed was not a pre-existing agency relationship (which was very doubtful), but the fact that the defendants had voluntarily assumed responsibility for the trust’s affairs. The reasoning of the courts leads to two problems: Rather than considering whether an implication of fiduciary duties is appropriate, the courts spend a lot of time and mental energy in deciding whether or not the relationship between the parties is properly characterized as agency, and then, having affirmed that question, impose very wide-reaching, arguably too wide-reaching, fiduciary duties.

The Content of Fiduciary Duties

As we have seen above, in the general law of contract, English law refuses to recognize that the parties should conduct themselves in a way that accords with a general duty of good faith towards one another. The reason why this is said is that the supposed duty is too diffuse, too poorly defined, so that it might give rise to a “palm tree,” i.e.,discretionary, justice. Yet when it comes to fiduciary duties imposed on agents, it seems to have no such qualms. Here, English law seems to start with a general principle and then derive from these specific rules that govern the relationship between principal and agent.At the bottom, these turn out not to be so very different from the more precise rules set out in the ChineseCiνil Code.

Obligation to Aνoid Conflicts of Interest

This duty has two facets, both of which lie at the heart of the formulation in Art. 168 of the ChineseCiνil Code: The agent must not put himself in a position in which his selfinterest conflicts with that of the principal, and he must not put himself in a position in which the duties he owes to other parties, commonly other principals, conflict with the duties he owes to his principal.①Swain ν The Law Society [1982] 1 WLR 17, 36.Art. 168 focuses, rather more narrowly, on self-dealing and cross-dealing—the duty in English law is wider but Art. 168 is certainly capable of being interpreted analogously.

Obligation Not to Profit

Boardman ν Phippsitself②[1967] 2 AC 46.is the best example of this obligation and, at the same time,demonstrates its breadth. The defendants would not have come across the opportunity they took advantage of were it not for their involvement with the affairs of the trust, and that causal link alone sufficed to make them disgorge them to the trust, which thereby received a windfall.

One way to look at this, of course, is as a sub-rule of the previous rule, to avoid conflicts of interest. But I would suggest it takes that rule very far indeed. After all,the trust would not have had the means to pursue the opportunity the defendants had identified—the prophylaxis against conflicts of interests, even of a theoretical nature, is a worthy goal, but it is very arguable that this is one example of it going too far.Boardman ν Phippsis very far removed from the clearer case of profiting from one’s agency, namely situations in which the agent takes a bribe or makes some other form of “secret” profit.Where an agent, having to decide to which the third party to commit his principal, takes a bribe from one of the third parties, it is very arguable that the amount of the bribe would have accrued to the principal had the agent not been dishonest.

Obligation to Act in the Beneficiary’s Best Interests

Professor DeMott, now the reporter of the American Restatement of Agency (3d),suggests that “the fiduciary’s duties go beyond mere fairness and honesty; they oblige him to act to further the beneficiary’s interests” (DeMott, 1988, pp. 879, 882). It is surprising to find such a vague duty in a system that eschews the recognition of a general duty to act in good faith precisely because it considers it too vague! And, indeed, unused to general principles, the application of this duty has led to surprising results. Thus, inFassihi ν Item Software (UK) Ltd,③Fassihi ν Item Software (UK) Ltd [2004] EWCA Civ 1244; [2004] BCC 994, cited by James Edelman, When Do Fiduciary Duties Arise?, 126 Law Quarterly Review, 302, 321 (2010). Wallace ν United Grain Growers Ltd [1997] 3 SCR 701 at [98].the Court of Appeal decided that it was in the best interests of a company for a director to disclose his own wrongdoing—a decision which contradicted the well-established House of Lords decision inBell ν Leνer Bros,①Bell ν Leνer Bros [1932] AC 161.fundamental to the law relating to a mistake in contract, which held the opposite, namely that a company director did not have to disclose his own breaches of duty. Lord Atkin had, in that case,argued that the adversarial principle prevailed in such a case, arguing that “to imply such a duty would be a departure from the well-established usage of mankind and would be to create obligations entirely outside the normal contemplation of the parties concerned.”②Bell ν Leνer Bros [1932] AC 161, 228.

Obligations to Act in Good Faith

What the courts mean when they refer to this fiduciary duty is more than a mere duty of honesty (which is well-established to exist throughout the law). What precisely is meant by it, beyond the three more (or less) concrete duties listed above, is not entirely clear. It has indeed been described as “incapable of precise definition.”③Fassihi ν Item Software (UK) Ltd [2004] EWCA Civ 1244; [2004] BCC 994, cited by James Edelman, When Do Fiduciary Duties Arise?, 126 Law Quarterly Review, 302, 321 (2010). Wallace ν United Grain Growers Ltd [1997] 3 SCR 701 at [98].

It is striking that it is this precise duty of good faith which English law refuses to recognize as a general principle. Professor Edelman argues that this is, in fact, not correct and that English law does recognize such a principle, but one which will manifest itself in different ways depending on the situation to which it is argued to apply (Edelman, 2010,pp. 302, 324). This is in keeping with his general argument by which fiduciary duties are implied in the agent’s undertaking to the principal, to no greater or lesser extent than is necessary, to make the relationship work. While agreeing with the general argument,I would argue that the inclusion of the very wide and diffuse obligation to act in good faith within the fiduciary obligations attributed to agents is simply a recognition that, in contrast to the general law of contract, the principal/agent relationship is not adversarial,but collaborative.

Conclusion

I have sought to show in this article that English law, once it switches from adversarial to collaborative relationships, imposes greater (and less certain) duties than would be required in a legal system that takes a less adversarial stance throughout. A system like the Chinese law of contract, which expects the parties to collaborate more rather than simply pursue their own self-interests without regard to the interests of their counterparties, is therefore not obliged to follow the lead of English law in imposing very wide-ranging and ill-defined fiduciary duties.