‘Agency Problem’ Cases
5
‘Agency Problem’ Cases
I Introduction
This chapter places cases involving concurrent breaches of fiduciary duty and breaches of contract and cases involving a breach of a negative covenant within the same category, under the umbrella of ‘agency problem cases’. Both involve the same basic scenario. Boris contracts with Alice not to do a specific thing or things which relate to Alice’s best interests, but Boris then breaches the contract and goes ahead and does the very thing which he has contracted not to do, and profits thereby. In the case of fiduciary duties Boris is obliged not to conflict with Alice’s interests and not to make an unauthorised or secret profit. In the case of a negative covenant, boris’ obligation is not to do the particular thing specified in the negative covenant, but in addition, it will be argued that where trust is conferred on Boris, he also has an obligation not to profit, and if he does profit, that profit should be stripped from him. It should be noted that I do not include those cases which deal with negative covenants providing that the defendant must not provide his services to anyone other than the claimant within the category of ‘agency problem cases’.1
Both breach of fiduciary duty and breach of negative covenants raise what Professors Thel and Siegelman call ‘the agency problem’.2 ‘Agency’ in this context does not refer to the legal concept of agency, but the economic concept of ‘principal– agent’, which attempts to ascertain how a promisee (principal) can design a contract which motivates another individual, the promisor (agent) to act in the promisee’s interests.3 An ‘agency problem’ arises when the promisee has less information than the promisor about the actions the promisor either has undertaken or should undertake (‘information asymmetry’). Agency problems can create a situation of ‘moral hazard’ because the party that is insulated from risk generally has more information about its actions and intentions than the party paying for the negative consequences of the risk. In both fiduciary duty and negative covenant cases the promisee has particular difficulty in monitoring the promisor’s performance as a result of information asymmetry. The promisor can breach before the promisee has a chance to apply for an injunction, and the breach cannot be undone. The promisor is not the one who suffers from the breach; indeed in these cases he profits as a result. The promisee depends upon the good faith of the promisor. Therefore, in order to address the agency problem, the law needs to provide incentives for the promisor (agent) to behave in accordance with the promisee’s (principal’s) wishes. Otherwise the promisee’s performance interest is not adequately vindicated by the available legal remedies.
An agency problem involving a negative covenant would arise if you promised me that you would not write a book detailing my innermost secrets, and I paid you $1000 for this. In breach of your promise, you wrote a book detailing my secrets and received $100,000 from the publishers as an advance.4 I did not find out about your book until it had been released and it was too late to obtain an injunction to stop you publishing. In such a case, I do not have the relevant information about your intentions in relation to performing the contract. You do not have to bear the negative consequences of breaching the contract; indeed the consequences for you are positive. I cannot prevent the breach which has already occurred, although perhaps I can get an injunction to prevent ongoing breaches. Any compensatory damages for which you are liable will be far outstripped by your profit, and in any case, it is difficult to see any pecuniary loss on my part which can be rectified by monetary compensation. However, if you knew that you might well be legally required to disgorge the profits you had made in breach of contract, you would either perform your obligation not to publish or negotiate a release from the contract, because there would no longer be any incentive to breach unilaterally. Your profit would be stripped from you. Disgorgement damages thus help to rectify my lack of information about your intentions. The deterrent rationale of disgorgement damages operates strongly in this situation because the aim is to deter unilateral breaches and provide incentives for performance or negotiation instead.
The agency problem in breach of fiduciary duty cases has been solved by the availability of profit-stripping remedies which remove the promisor’s incentive to breach.5 It is uncontroversial that disgorgement damages are available for breach of fiduciary duty and for other equitable wrongs such as equitable breach of confidence. By contrast, the law of contract generally does not presently provide incentives for a promisor in a negative covenant situation to perform the promise or, alternatively, to negotiate with the promisee for a release. If compensatory damages are inadequate and specific performance is no longer available, the promisor has an opportunity to breach which he should not have. The promisee’s interest in performance is not adequately vindicated by the courts, and expectation damages do not adequately guard against the breach of trust in such a case. Courts rarely explicitly award disgorgement for breach of a negative covenant. Attorney-General v Blake is the watershed case in which disgorgement damages were openly awarded, when – as discussed in chapter three – George Blake, a former spy published a book detailing his double agency for the Soviet Union and profited thereby, breaching a negative covenant in his contract of employment by failing to consult the British Government about his publication even though his employment had long since ceased.6
What then of substitutability, which played such a prominent part in the justification of disgorgement damages in the previous chapter? Substitutability remains relevant as to why specific relief and disgorgement damages are awarded for concurrent breach of fiduciary duty and contract and breach of negative covenant. The rights granted by such contracts are intrinsically non-substitutable because they are intangible and cannot be bought or sold in any market.7 In addition, substitutability requires us to consider the idiosyncratic value of the promise to the claimant. Substitutability is particularly useful in breach of negative covenant cases where disgorgement is awarded as a surrogate for expectation damages. 8 When the defendant has benefited in a measurable way and has caused a loss which is speculative or hard to quantify, it will be in the claimant’s interests to obtain disgorgement damages.9 The kinds of case in which it is typically difficult to measure the losses of the claimant are often intellectual property infringement cases, in which it is very difficult to ascertain what the losses of the claimant are as a result of the infringement, but the sales of the defendant are clear and quantifiable. Nonetheless, substitutability does not provide a complete explanation of the outcome of the negative covenant cases. It is only the first step. If substitutability was the sole criterion, then every breach of negative covenant would give rise to disgorgement damages.10 Something more is needed to explain why damages are inadequate.
As with the ‘second sale’ cases, in order for a defendant to be liable for disgorgement damages for breach of negative covenant, the breach must be advertent. In breach of fiduciary duty cases, courts are willing to impose liability on parties whose breaches are bona fide, and thus the punitive justification does not sit so well with fiduciary duties. These cases are better justified purely by reference to the deterrent intention.
This chapter will first explore the core uncontroversial cases for disgorgement damages which involve concurrent breach of fiduciary duty and contract. It will then suggest that cases for breach of negative covenant lie on the margins of fiduciary and contract law. Some cases fall within the fiduciary realm (although the justification for their doing so is not always convincing) and others do not. Disgorgement damages are increasingly available not only for those cases which fall within the fiduciary realm, but to those ‘quasi-fiduciary’ contracts on the margins which suffer from the ‘agency problem’. A negative covenant does not necessarily create a fiduciary duty, but it can approach the margins of fiduciary law, particularly when the contract in question is not designed to make a profit, but rather to protect some other kind of interest (including contracts to protect national security, public interest, to resolve a legal dispute or to protect one’s family). If the defendant breaches the contract without the claimant knowing, then there is very little the claimant can do about it. The damage has been done, and the clock cannot be turned back. Therefore it is necessary to have some kind of deterrent remedy to prevent the breach occurring in the first place.
II The ‘Agency Problem’ and Disgorgement
The core ‘agency problem’ cases involve concurrent breach of fiduciary duty and contract. In these cases, the principal has difficulty in monitoring the performance of the fiduciary on whom power is conferred. In this section, I will first consider the uncontroversial availability of disgorgement for breach of fiduciary duty and then consider the availability of disgorgement for breach of negative covenant. I will suggest that, where courts have wished to award disgorgement damages because of the ‘agency problem’, they have used the fiduciary concept in an instrumental fashion which is not wholly convincing.
A Disgorgement, the ‘Agency Problem’ and Fiduciary Duties
Fiduciary duties are often set apart from other legal duties (including those arising via contract, tort and unjust enrichment). Cardozo J described the duties of a fiduciary in Meinhard v Salmon:
Many forms of conduct permissible in a workaday world for those acting at arm’s length, are forbidden to those bound by fiduciary ties. A trustee is held to something stricter than the morals of the market place. Not honesty alone, but the punctilio of an honor the most sensitive is then the standard of behaviour. As to this there has developed a tradition that is unbending and inveterate. Uncompromising rigidity has been the attitude of courts of equity when petitioned to undermine the rule of undivided loyalty by the ‘disintegrating erosion’ of particular exceptions. Only thus has the level of conduct for fiduciaries been kept at a level higher than that trodden by the crowd. It will not consciously be lowered by any judgment of this court. [Emphasis added]13
Typically, the fiduciary obligation is imposed by reference to the particular relationship between the parties or their status. There are certain relationships which are presumed to be fiduciary (eg trustee and beneficiary, solicitor and client, company and director, agent and principal).
Let us take, for example, one of the ‘inner circle’ of fiduciary relationships, the relationship between solicitor and client,14 as an illustration of how the ‘agency problem’ arises. When the client consults a solicitor, the client is entitled to assume that the solicitor will not give her self-interested advice (or advice which preferences another client over this client) and that the solicitor will not make a profit by breaching his duty towards her. The client is not well situated to assess whether or not the solicitor is acting in her best interests. The solicitor exercises discretion over the client’s affairs, and consequently he may defalcate or make a profit by breaching his duty towards her without her knowledge. However, if any profit made in breach of a fiduciary duty is stripped, then there is no incentive for the solicitor to breach his duty towards the client. Professor Worthington has argued that disgorgement is peculiarly well-suited for preventing breaches of equitable obligations:
The aim is to exact particular standards of conduct in the protected relationships; to this end, the relevant law is concerned with proscribing certain activities, not with precluding particular outcomes. The appropriate remedial response for breaches of these equitable obligations is disgorgement because this is the remedy which best supports the legal obligation being enforced.15
A fiduciary duty may be imposed outside the ‘presumptive circle’, even in the context of a contractual relationship which is ordinarily self-interested. Usually, courts consider whether the fiduciary has voluntarily undertaken or agreed to act on behalf of another, such as Mason J’s ‘undertaking test’ in Hospital Products Ltd v United States Surgical Corporation.16 Alternatively, it is said that the fiduciary has the exercise of a power or discretion to which the beneficiary is vulnerable.17
Commentators have searched for a unifying principle of fiduciary relationships, but again, the cases are notoriously difficult to categorise, so that Sir Anthony Mason has said ‘the fiduciary relationship is a concept in search of a principle.’18
Justice Easterbrook and Professor Fischel have suggested that the difficulty in discovering a unifying principle of fiduciary relationships arises because scholars are mistakenly looking for something unique or special about fiduciary relationships. 19 They argue that, in fact, fiduciary relationships are just an extreme variety of contractual relationship,20 and that fiduciary relationships tend to arise when a transaction is very complex, and one party wishes to impart a discretion on the other, but it is too difficult to specifically enumerate each and every undertaking.21 Edelman, too, has argued that fiduciary obligations are just another species of voluntary undertaking, and it is for this reason that the scope of the fiduciary duty depends upon the circumstances and scope of the undertaking involved.22
Worthington has argued that accounts of profit are appropriate for fiduciary duties because the aim of the remedy is to ensure that the imposed obligation is not breached. This is in contrast to compensatory damages, which are typically the default remedy for non-equitable private law duties such as tort and contract, and simply seek to ensure that a breach does no harm.23 Where equitable duties are concerned, as previously mentioned, she argues that the law is concerned to prevent certain activities, not with preventing certain outcomes.24 I will argue that the same can hold true for some breaches of negative covenant, and thus disgorgement damages may be an appropriate remedy.
B Disgorgement, the ‘Agency Problem’ and Negative Covenants
While disgorgement damages are an uncontroversial remedy for breach of fiduciary duty, until Blake at least, they were not awarded for breach of negative covenant in the absence of a fiduciary duty. Generally speaking courts are far more willing to award disgorgement for breaches of equitable obligations than for breaches of common law obligations.25
By contrast, courts are prepared to award an injunction to restrain a breach of negative covenant where specific performance of the contract as a whole would not be ordered (known as an injunction pursuant to the rule in Lumley v Wagner).26 As with specific performance, the basis for the injunction is that ‘damages are inadequate’ or that it is ‘necessary to do justice’. It has sometimes been said that injunctions for a breach of negative covenant in a contract will be granted as a matter of course, drawing on the following statement of Lord Cairns in Doherty v Allman:27
My Lords, if there had been a negative covenant, I apprehend, according to well-settled practice, a Court of Equity would have had no discretion to exercise. If parties, for valuable consideration, with their eyes open, contract that a particular thing shall not be done, all that a Court of Equity has to do is to say, by way of injunction, that which the parties have already said by way of covenant, that the thing shall not be done; and in such case the injunction does nothing more than give the sanction of the process of the Court to that which already is the contract between the parties. It is not then a question of the balance of convenience or inconvenience, or of the amount of damage or of injury – it is the specific performance, by the Court, of that negative bargain which parties have made, with their eyes open, between themselves.28
However, despite the apparent adoption of this dictum in some cases,29 the bulk of the case law suggests that the usual equitable discretions remain relevant.30 Nonetheless, courts are far more ready to award injunctions restraining the breach of a negative covenant than they are to enforce a positive obligation. It has been speculated that this is because injunctions do not contradict the claimant’s duty to mitigate, they do not infringe individual liberty to the same extent as a positive order, they do not have the problem of constant supervision, and damages are often less easy to assess for breach of a negative covenant compared to a positive obligation.31
Many of these cases involve a covenant that property will not be used for certain purposes or in certain ways.32 Or there may be a clause in a contract to the effect that the defendant will only supply products to the claimant33 or use products supplied by the claimant.34 Although it was an employment contract, the contract in Blake falls into this category, because the clause did not seek to restrict the provision of George Blake’s services. Instead, Blake promised not to publish any information obtained in the course of his employment without seeking the consent of his employer first.
As noted previously, the reason why specific relief is granted is partially as a result of lack of substitutability. The rights granted by the negative covenant cannot be bought or sold in the marketplace.35
However, in order to fully understand why specific relief is readily awarded for breach of negative covenant, we should also revisit Birks’ suggestion to have ‘regard to the objectives which the victim of the breach had hoped to achieve through performance of the contract.’36 This is the aspect of substitutability which looks to the idiosyncratic value of performance to the claimant. The courts look at what the claimant hoped to gain through performance of the contract, and ascertain whether compensatory damages would put the claimant in a position as if the contract had been performed. An injunction is readily awarded because once the negative covenant has been broken, it cannot be undone, or as Pomeroy puts it:
Judges have been brought to see and to acknowledge . . . that a remedy which prevents a threatened wrong is in its essential nature better than a remedy which permits the wrong to be done, and then attempts to pay for it37
An injunction is intended to prevent future wrongdoing or to ameliorate future consequences of a past wrong.38 Without the security provided by an injunction, the defendant’s bargain will be valueless to the claimant. The claimant is at the mercy of the defendant, and cannot rely on anyone else to perform that agreement. Nevertheless, there may be circumstances where an injunction is not granted for discretionary reasons (hardship, lack of clean hands and the like).
Breaches of negative covenant usually involve a specific promise not to do a particular thing. Unlike a fiduciary, a contractor who is subject to a negative covenant does not have discretion. The choice is to perform the obligation, or to fail to perform the obligation. There is no obligation on the part of a promisor to act in the best interests of the promisee.
Courts do award disgorgement damages for breach of negative covenant, but, at least until Blake, they seldom do so openly. The tendency before Blake was to characterise the contract containing the negative covenant as ‘fiduciary’ in an instrumental fashion, purely in order to obtain the disgorgement remedy. Courts did this because they instinctively sensed that the claimant’s right to performance should be protected and that the defendant should be deterred from breaching the negative covenant because of the ‘agency problem’.
It is trite to say that the fiduciary concept has often been misused in order to achieve a just result.39 The label ‘fiduciary’ is notoriously flexible, and has been dubbed the ‘accordion term’ because of its ability to be stretched or contracted.40 There is a risk that the term will be rendered meaningless if it is stretched too far merely because a court wishes to award a particular remedy. Some of the cases cited by Lord Nicholls in Blake as examples of cases where disgorgement damages have been awarded for breach of fiduciary duty concurrent with breach of contract involve instrumental applications of the fiduciary concept.41 An instrumental use of the fiduciary concept means the judicial determinations lack transparency; the real reason for the decision is concealed.
One of the cases relied upon by the House of Lords in Blake was Snepp. There, the US Supreme Court found that Snepp, a former employee of the Central Intelligence Agency (‘CIA’) was liable to disgorge profits made after he released an unauthorised book about the activities of the US Government and the CIA in Vietnam. Snepp was found to owe concurrent contractual and fiduciary obligations to the CIA, and the CIA was awarded a constructive trust over the profits. The US Supreme Court subjected Snepp to a fiduciary obligation because of the ‘extremely high degree of trust’ involved in his relationship with the CIA and the US Government.42 However, as noted by the English Court of Appeal in Blake, it seems extraordinary to subject Snepp to the full brunt of fiduciary obligations even after his employment had finished.
Easterbrook and Fischel observe that negative covenants are often included in contracts where one party seeks to govern the future conduct of the other, but cannot specify exactly which obligations will arise in advance without costly and inconvenient negotiation.43 They regard the negative covenant in Snepp as initiating a process for making decisions in the future as to what Snepp could and could not write. Snepp should have tried to negotiate a second contract to ascertain exactly which parts of his manuscript could be published:
The first contract established the employment relation and the submission requirement. The second contract would establish the terms of publication. Just as restitution plus an additional penalty induces the would-be thief to enter into market transactions instead, the profits remedy induces the parties to contract explicitly. It is a contract inducing, not a contract-frustrating, approach.44
The threat of a disgorgement remedy removes the incentive for the promisor to breach, and instead gives the promisor incentive to negotiate with the promisee or to perform the contract. The rationale behind this is deterrence; otherwise negative covenants become ineffectual. Where negative covenants have been breached, it is too late to deter the promisor in relation to the breach which has already occurred, but the promisee’s right is vindicated by stripping the profit from the promisor. In those cases where disgorgement has been ordered, the concern of the courts is more to exact adherence to the specific obligation rather than to ensure that the breach does no harm. In most cases, compensatory damages are wholly inadequate to recognise the claimant’s performance interest, and the law ought to attempt to prevent potential defendants from breaching in the first place rather than compensate for harm once the breach has occurred. Once the breach has occurred, the promisor’s profit is stripped in order to vindicate the promisee’s right to performance and in order to punish the promisor.
In analysing the overlap between the breach of fiduciary duty cases and the breach of negative covenant cases, I propose that fully-blown fiduciary duties have a special ‘bundle of obligations’ attached to them. These obligations can be limited or shaped by contract. Some of the negative covenant contracts have a modified, less demanding version of the fiduciary ‘bundle of obligations’ implied into them. Crucially, the negative covenant cases include a duty to act in good faith which distinguishes them from ordinary commercial contracts, in part because of the ‘agency problem’ which also bedevils fiduciary law. This explains why breach of negative covenant sits on the margins of fiduciary law, and why disgorgement damages are available for some breaches of negative covenant. They could be said to be ‘quasi-fiduciary’ cases, falling between fully-blown fiduciary obligations and purely self interested contract law.
III The Fiduciary ‘Bundle of Obligations’
Fiduciary relationships generally have a special ‘bundle of obligations’ attached to them. Fiduciary obligations can be limited and shaped by the terms of a contract,45 and accordingly, the line between contract and fiduciary obligation is not sharp. The breach of negative covenant cases are on the margins of fiduciary law because they also raise the ‘agency problem’ and thus they may also be candidates for awards of disgorgement damages, even if no fiduciary relationship is found. Courts do not always require a fiduciary relationship when imposing remedies which are usually an incident of a fiduciary relationship.46 For example, in White City Tennis Club a limited fiduciary duty was found on the facts, but Macfarlan JA would still have awarded a constructive trust even if one had not been established. 47 A fiduciary relationship was not a necessary precondition for an award of a constructive trust.48 Indeed, the line of authority represented by Muschinski v Dodds49 and Baumgartner v Baumgartner50 involves imposing a constructive trust in situations where no fiduciary relationship exists.
There is a particular group of duties which are typically characterised as fiduciary:
1. The fiduciary must not place himself in a position of conflict between his own interests and those of the beneficiary without consent, nor must a fiduciary allow his duty towards two different beneficiaries to conflict.
2. The fiduciary must not make an unauthorised or secret profit from his fiduciary position.
3. The fiduciary must act in the best interests of the beneficiary, that is, an obligation to act for the proper purposes for which the fiduciary has agreed to act.
4. The fiduciary must act in good faith, that is, act in a manner which is candid, honest, reasonable and forthright.51
I call these obligations the ‘fiduciary bundle of obligations’. It will be suggested that different voluntary undertakings may have all, some or none of the ‘fiduciary bundle of obligations’ attached to them.
A party who is subject to a ‘fully-blown’ fiduciary obligation will be subject to all four of the fiduciary bundle of obligations. But there are other voluntary obligations to which only some of the fiduciary bundle of obligations may attach or, most frequently, none will attach. It is worth noting that some of the specific fiduciary obligations can be excluded by an appropriately drafted clause in a contract. 52 The only obligation which cannot be excluded by contract is the duty of good faith.53 In addition, even in a fiduciary context, the full ‘fiduciary bundle of obligations’ may not be present depending upon the circumstances. In Kelly v Cooper54 the Privy Council suggested that fiduciary obligations may be limited by the nature of the activities pursued by the fiduciary. The estate agent in that case owed fiduciary duties to his clients, but the nature of these obligations was modified by the fact that he worked for multiple principals. Thus, the estate agent did not have to disclose confidential information about other clients to the claimant.
If we take the ‘fiduciary bundle of obligations’ above and see how they apply to the negative covenant cases, including those cases which are arguably masquerading as fiduciary, it is hard to see that there is a fiduciary-type obligation to act in the best interests of the other party. An obligation to act in the best interests of another party suggests that the promisor has discretion to use a particular power for proper purposes. However, many of the negative covenant cases do not involve discretion at all. The obligation is unequivocal: the promisor is not to do the stipulated thing. The only discretion on the part of the promisor is whether the promisor chooses to breach.
Similarly, the duty not to conflict is extremely limited in some of the cases involving negative covenants and concurrent fiduciary duties. By contrast, the notion of conflict of interest is extremely broad in many ‘proper fiduciary’ cases. So the majority in Boardman v Phipps found that there was no need for an actual conflict of interest: ‘even if the possibility of conflict is present between personal interest and the fiduciary position the rule of equity must be applied.’55
One of the suggestions made by the Court of Appeal in Blake was that disgorgement damages should be awarded where the defendant had done ‘the very thing which he agreed not to do’.56 This was rejected by Lord Nicholls in Blake.57 However, a conflict of interest arises in these negative covenant cases in a very limited sense, and it arises precisely because the defendant has done the very thing which he agreed not to do. The defendant agreed to do something for the benefit of the claimant, and if the defendant had kept his obligations, he would have foregone a potential profit. By doing the very thing he had agreed not to do, and making a profit thereby, the defendant should be obliged to disgorge that profit. There will be a conflict of interest in these cases.
To summarise, the primary obligations in negative covenant cases are a modified and less stringent version of the ‘fiduciary bundle of obligations’. The obligations typically include the following: