Definitions and Controversies
TO BEGIN TO investigate ‘restitution’ and ‘public bodies’ the starting point must be a definition of those terms. It is necessary, therefore, briefly to outline the relevant matters from each sphere in order to provide a secure foundation for the specific investigation which will combine the two.
The first important thing to note is that, like the law of obligations more generally, this area has Roman origins. The Digest contains two versions of a fragment excerpted from Pomponius. One reads ‘this is indeed by nature fair, that nobody should be made richer through loss to another’.1 The other reads ‘[i]t is fair by the law of nature that nobody should be made richer through loss and wrong to another.2 In Gaius’ Institutes, the law of obligations was divided by the events that brought the obligations into being. Gaius, unable to fit the case of a reclaimed mistaken payment into his statement that all obligations arose from contract and tort,3 listed another category: other miscellaneous events. Either he, or ‘someone later interfering with his text’,4 then broke this third category into two, creating the categories of obligations arising quasi ex contractu and quasi ex maleficio,5 which were subsequently adopted by Justinian.6 In this context, the Latin term was less inappropriate than the common law term quasi-contract: the Roman category included all obligations which were not contracts, but whose consequences were similar to the results of a contract. It could thus genuinely be said that they arose ‘as if upon a contract’, a term that made it perfectly clear that there was no actual contract. It is important for our purposes to realise that these Roman origins are the starting point both for the development of English law7 and of our comparator system, French law. However, whereas in France this area of law remained separate from other branches of the law of obligations, in English common law the terminological connection with the law of contract for many years deprived the subject of an independent existence and of a reasoned basis.
The historical confusion surrounding the subject has been well documented.8 Its result was that in England the subject did not begin to develop as a discrete area of the law until the demise of the ‘implied contract’ theory in the twentieth century. This process has often entailed reinterpretation of rules and decisions previously thought to belong to a variety of disparate areas of the law. Consequently, there is still great controversy over the extent to which this process of gathering and reinterpretation should continue and thus over the content and reach of the area itself.9
One example of this controversy concerns the very name given to the area. The first major gathering exercise was undertaken by Goff and Jones in 1966, but one of the most important steps in the reinterpretation exercise in English law came from Birks’ An Introduction to the Law of Restitution in 1985. He began by pointing out that from a theoretical perspective ‘restitution’ cannot be aligned with ‘contract’ and ‘tort’ because it does not denote an ‘event’, but rather the response to an event.10 We do not refer to tort as ‘compensation’ or to contract as ‘expectation damages’, so although thus far the terms ‘unjust enrichment’ and ‘restitution’ have been used without specific import, from now on the relevant area will be referred to as ‘unjust enrichment’ rather than ‘restitution’. Whatever the other problems concerning the relationship between the two terms, it is this initial event with which we will be concerned.
A second instance of controversy concerns the technique used, particularly by Birks, to explain and analyse unjust enrichment. This is the technique of ‘mapping’ or ‘topography’. He argues that ‘case-law grows without much regard for principle or for the coherence of one piece of law with another’11 and, in particular, he sees it as the role of academic investigations to find a ‘line of best fit’ through the existing law. However, as he himself notes, such an exercise is not purely descriptive, but also contains normative elements: ‘[a] skeleton of principle is not just the common sense behind a legal topic, but rather a particular organisation of its common sense, a version chosen from a number of possibilities’.12 Thus, when the cases do ‘reflect back the framework selected as best-fitted to the subject’, it is a framework which contains conclusions about how best the law should be seen.13 For this reason, however, others see the map as ‘an attempt to squeeze the law into a mould it patently does not fit’.14 It is, therefore, to a certain extent, impossible to separate the analytical technique used in such an investigation from the controversy underlying the area as a whole. This investigation will use the existing topography as a starting point, because it will be seen that the choices made by the process so far present a helpful way of highlighting the relationship between the two areas of law, public law and unjust enrichment, brought together in this study. However, this does not mean that the mapping process can by itself provide enough normative justifications for the courts to answer the questions before them in public body cases. Nor does it mean that those cases will automatically be fitted into the existing map without independent justification. Like any map, this one will provide a basic structure to prevent the investigation from becoming lost, but it will not by itself dictate the route to be followed in future.
The first thing to note is that if the law is to be divided up, or classified into various ‘territories’, then there is a variety of possible means of categorisation. Moriarty lists several: Birks’ system is known as ‘event-response’ classification, but it would also be possible to divide the law up by the subject-matter of the obligation (money, goods, work) or by social context (family law, company law, labour law) as happens in universities. The point is, Moriarty argues, that since tort and contract already follow the event-response classification it is this one we should use to discuss unjust enrichment.15
This is particularly important in the present context, since the current investigation seems to span two such types of classification: private law questions relating to causes of action in unjust enrichment which belong to the event-response Birksian classification, and the relationship between public and private law, which seems initially more like the ‘social context’ classificatory system.
Remaining for the moment with the event-response classification, Birks has developed this as follows: ‘All rights are either (by jurisdictional origin) legal or equitable and (by the nature of their exigibility) proprietary or personal; and all such rights arise from wrongs, from consent, from unjust enrichment or from other causative events’.16
It is true that often the law does not use the language of ‘events’ and ‘responses’, preferring the terms ‘right’ and ‘remedy’. However, in his lecture entitled ‘Rights, Wrongs and Remedies’ (delivered as the 23rd Blackstone Lecture in May 1999),17 Birks defends the event-response classification from the threat posed by the word ‘remedy’ by abandoning this word and concentrating instead on the word ‘right’ since this will allow us to see that the courts ‘realise rights not only from wrongs, but also from not-wrongs, and, more exactly, from three categories of not-wrongs’.18 This will mean in turn that we can focus properly on the causative event relied on by the claimant, and ensure that the law responds correctly.
If the law is to be divided by causative events into consent, wrongs, unjust enrichment and ‘others’, it becomes necessary at the next level of detail to examine the boundaries within each of those categories. In the category entitled ‘unjust enrichment’ we are again immediately confronted by further controversy. In his Introduction, Birks argued that the law of unjust enrichment was divided into two. Sometimes the enrichment was at the claimant’s expense because there had been a direct transfer of value from C to D, for example where C pays D £100 by mistake. This became known as ‘autonomous unjust enrichment’ because it was totally independent of all other events. On the other occasions D wronged C and in doing so made a profit, for example where X pays D £100 to hit C, and Birks called these cases ‘enrichment by doing wrong’.19 However, in his more recent work Birks concluded that in the wrongdoing cases the operative event is the wrong itself.20 Thus for Birks, the boundary is no longer one within unjust enrichment, but one which defines the edges of the subject. This also provides one reason why, even after the event-response distinction is made, for Birks the response of restitution no longer quadrates with the event of unjust enrichment.21 It should not be thought that this is only of semantic interest. If the event in question is a wrong, then policy arguments from the law of wrongs will be allowed to limit recovery. If, on the other hand, the relevant event is an unjust enrichment, then it is policy arguments from that sphere which must operate. The controversy arises because Birks’ new viewpoint has not been universally accepted. On the contrary, it has been challenged descriptively by Friedmann,22 normatively by Tettenborn23 and Kull,24 and at least temporarily by Burrows.25 This debate has the potential to affect the current investigation in various ways. The standard examples of overpaid tax and money transferred under contracts that public bodies had no power to form seem initially to fall into the ‘autonomous unjust enrichment’ category. On the other hand, since these charges or payments are unlawful as a matter of public law, if they are not to be placed alongside cases where money is gained through a wrong, this decision will require justification.26 Even if we conclude that they do not, it will of course remain possible that there are other situations in which a public body may gain money through wrongdoing.27
It is generally accepted28 by supporters of the unjust enrichment principle that four elements must be fulfilled in order for a claim in unjust enrichment to succeed:
1. The defendant must have been enriched.
2. This enrichment must have been received at the expense of the claimant.
3. The enrichment must be unjust.
4. There must be no defences.
Two of these criteria are particularly relevant to this investigation.
The Enrichment Must Be Unjust
It is on this question that much of the investigation will be based since it is arguable that this element is the most important factor in explaining the law’s intervention. For example, this is the element which has, at least in the past, contained the greatest difference between common law and civil law systems such as in France and it is this element on which some of the most recent controversy has focused. In Orakpo v Manson Investments Ltd Lord Diplock said that:
[T]here is no general doctrine of unjust enrichment recognised in English law. What it does is to provide specific remedies in . . . cases of what might be classified as unjust enrichment in a legal system that is based upon the civil law.29
In his Introduction to the Law of Restitution, Birks made it clear that ‘nothing’ in that book ‘is meant to contradict’ Lord Diplock’s position, but he also argued that such a general doctrine would be ‘unusably vague’30 and he was therefore concerned to find a better order for the specific instances Lord Diplock did recognise. Thus the burden was to rest with the claimant of proving a reason for restitution, an ‘unjust factor’31 to show why the defendant should have to give up the enrichment. It would not be open to the claimant, as it is in civilian systems, to claim that the enrichment was without cause and thus to place the burden of proving a legal cause on the defendant. Clearly, when he was writing his Introduction, Birks’ view was that it was desirable that the burden should remain on the claimant, so that ‘unjust’ can be made to ‘look downwards to the cases’.32 For this reason, Birks argued, fears that the concept of ‘unjust enrichment’ will lead to abstract and uncontrollable ‘palm-tree’ justice33 are unfounded. The phrase is only used to identify the situations in which the law will respond to an enrichment of one person at another’s expense. If it is undesirable for the law to respond to this in an abstract way, it is perfectly possible to construct rules to ensure that it does not do so.
However others, and more recently Birks himself,34 have argued on the contrary that a generalised right to unjust enrichment is either preferable or necessary.35 It is therefore important to examine whether there is any practical difference between a generalised right concretised by specific rules and a right based only on specific causes of action, but justified by an underlying general principle. In Deutsche Morgan Grenfell v IRC,36 Lord Walker regarded English law as being ‘at something of a crossroads’37 between the two approaches, and indicated that his own ‘tentative inclination’ was to align English law with Scottish law and other civilian systems,38 but ultimately concluded that ‘the choice is one which will rarely make much if any practical difference to the outcome of any particular case before the court’.39 Burrows’ view is similar, especially since the English approach is not static and is capable of recognising new unjust factors,40 and even Birks himself accepts that even if ‘absence of basis’ is to be the final reason for recovery, beneath this apex of the pyramid ‘at the base, the particular unjust factors such as mistake, pressure, and undue influence become the reasons why, higher up, there is no basis for the defendant’s acquisition’.41
Indeed, as Lord Walker pointed out, ‘the recognition of “no basis” as a single unifying principle would preserve . . . the purity of the principle on which unjust enrichment is founded, without in any way removing . . . the need for careful analysis of the content of particular “unjust factors” such as mistake’.42
This issue is particularly important in relation to claims against and by public bodies in relation to ultra vires payments, as Deutsche Morgan Grenfell itself shows, since it will be seen that here the ‘unjust factor’ itself can appear to be simply an absence of cause, in the sense that there was no legal basis for the claim made by the public body and the claim was therefore beyond the body’s powers.43 The proper interpretation of this apparently civilian approach will be discussed further in the detailed examination of the English system, and will be seen to be an important stage in the argument made there.44 It also follows that this area in particular will benefit from a comparison with the French system, which does adopt the civilian approach, so that the focus there is on finding a cause to justify the enrichment, rather than finding a means of ‘unjustifying’ it.45
But in the meantime, since the common law approach has not yet adopted this generalised right to restitution and instead requires proof of specific unjust factors or reasons for restitution,46 it will be necessary to identify what that reason is in the case of ultra vires payments to and from public bodies.
There Must Be No Defences
This element is also important and it is often the availability of a particular defence that is the primary reason for classifying a particular claim as belonging to the law of unjust enrichment or not, since some defences may be confined to cases of unjust enrichment while others may apply more broadly. Many defences and other related arguments are (at least theoretically) possible to a claim in unjust enrichment: time limits, change of position, estoppel, bona fide purchase, impossibility of counter-restitution, submission to an honest claim, ultra vires, passing on, fiscal disruption, exhaustion of statutory mechanisms for recovery and prospective overruling. The role of each defence in relation to the cases under investigation here will be discussed when more is known about the operation of those claims and will be seen to be closely connected to identification of the correct unjust factor.
Controversies surrounding the meaning of the word ‘enriched’ will only occasionally be relevant to this investigation, which is primarily concerned with the payment of money to and by public bodies. The requirement that the enrichment must have been received at the expense of the claimant will be relevant only in relation to cases where the defendant argues that the claimant passed the expense on to third parties.47
Having thus laid the basic ground from the point of view of unjust enrichment, it is necessary next to give a similar outline of ‘public law’.
At the outset it should be noted that there are those who argue that such a discrete area either does not or should not exist. The starting point for claims involving public bodies in England is the Diceyan orthodoxy that even public officials should be liable in the ordinary way before the ‘ordinary’ courts of the land, and there are those who argue that this either is or should be the case.48 Certainly at a very abstract level, the law applicable to public bodies and to private parties may well share similar values49 and even in more specific areas the rules applicable to both types of entity may be the same.50 Nevertheless, whatever one’s view of the desirability or precise location of the public/private divide, for the following two reasons it seems impossible to deny that to some extent it exists.51 The first and most concrete reason is that public law contains ‘remedies’ that are not available in private litigation. These ‘prerogative remedies’52 must be sought through a separate public law procedure, an application for judicial review.53 Whatever one’s normative view of the distinction, then, it is apparent that in this very concrete sense the law currently does regard public bodies as being subject to different rules from those applicable to private parties. But in addition to this immediate distinction it appears that public bodies also have both more and less power than private parties. Public bodies may be empowered to act in a way which in many cases is only permissible because the actor is a public body, the Revenue’s54 ability to levy tax being a prime example of this. Conversely, it may be desirable to restrain public bodies from acting in certain ways without this behaviour giving rise to any private law causes of action.55 This means that public law must contain events in addition to those on the private law map to which the law must respond but which do not fit easily into the private law categories.
Administrative law does not generally regard itself as charting a ‘taxonomy’ of these ‘events’ in the manner discussed above in relation to private law. Indeed, the concrete manifestation of the public/private divide has, at least in the past,56 centred around the existence of a separate procedure for judicial review, and even now accounts of public law generally see the ‘nucleus’ of the subject as being based around the public law ‘remedies’ now known as ‘quashing orders’, ‘prohibiting orders’, ‘mandatory orders’, together with at least declarations, injunctions and damages.57 This contrasts with the private law focus on events and responses discussed above. On the other hand, although it is clear that such a taxonomy of public law would look very different from anything found in the private law sphere, administrative law does essentially perform the same function. It certainly seems possible to define a biased decision as an ‘event’ to which the law responds, for example, with a quashing order. We are now not very far removed from public law’s second controversial debate over the role of the ‘ultra vires’ doctrine,58