AGGRAVATED, EXEMPLARY AND RESTITUTIONARY DAMAGES [1997] EWLC 247(5

BAILII is celebrating 24 years of free online access to the law! Would you consider making a contribution?
No donation is too small. If every visitor before 31 December gives just £1, it will have a significant impact on BAILII's ability to continue providing free access to the law.
Thank you very much for your support!

5. ADDITIONAL REFORM ISSUES AND PROPOSALS

5.1 A significant number of additional reform issues arise. In dealing with them we confirm our provisional view, expressed in the Consultation Paper and agreed with by all consultees who responded on the point, that a detailed legislative scheme, codifying the law on exemplary damages, should be attempted.(1)

5.2 It is important to recognise that the codification which we propose is codification only in the sense that it places punitive damages on a statutorybasis, and defines the most important characteristics and incidents of the remedy. The scheme is not intended to be an exhaustive statement of the applicable law. There are a significant number of rules which affect claims to exemplary or punitive damages which we do not seek to amend, which should continue to apply to punitive damages, but which we do not think it appropriate to state or refer to in our statutory scheme. Rules which fall into this category include: rules defining any wrong which founds a claim to punitive damages; rules relating to a persons capacity to be sued or to sue for a wrong;(2) rules on limitation of actions;(3) rules relating to the discharge of wrongs;(4) rules about a persons ability to limit or exclude his or her liability;(5) rules of private international law;(6) rules about assignment of claims to damages;(7) rules governing the awarding of interest on,(8) and taxation of,(9) awards of damages; rules relating to the deductibility of damages paid when calculating taxable profits;(10) rules relating to the ability of a person to prove an unliquidated claim to, or judgment debt for, damages upon the insolvency of a wrongdoer,(11) and as to the ranking of such a claim or debt.(12)

5.3 It is also important to emphasise at this point that nothing in our Act should be construed as stopping further common law development of the law relating to punitive damages, to the extent that such development would be consistent with our draft Bill. Nor should our draft Bill be construed as taking a particular view of the principles which currently apply at common law. For example, the fact that we have chosen to state expressly in a statutory provision that insurance against the risk of liability to punitive damages is not contrary to public policy, should not be taken to cast doubt on whether that may or may not be the current position at common law.(13)

(24) our draft Bill should lay down (in some instances by amending, and in other instances by restating previous law) the main elements of the remedy of punitive damages; but subject to this, the law relating to punitive damages should continue to apply and be open to future common law or statutory development. (Draft Bill, clause 1(1))

(1) The pleading of punitive damages

5.5 We consider the existing approach to the pleading of exemplary (or, as we prefer to call them, punitive) damages(14) to be the appropriate one. A claim to punitive damages should be specifically pleaded, together with the facts on which the party pleading them relies. Neither the court of its own motion, nor any other person or body, should be entitled to raise the issue if the plaintiff does not do so. And a plaintiffs failure specifically to plead punitive damages should prevent such an award from being made. The reason is that suggested in Broome v Cassell by Lord Hailsham:

. .. a defendant against whom a claim of this kind is made ought not to be taken by surprise.(15)

(25) punitive damages should not be awarded unless they have been specifically pleaded by the plaintiff, together with the facts on which the party pleading them relies. (Draft Bill, clause 3(2))

(2) The relevance of the means of the defendant

(a) How should the defendants wealth be relevant?

5.7 Inevitably the wealth of a particular defendant must significantly affect the extent of the punitive and deterrent impact of a punitive award. Nevertheless, we do not support inquiry into the financial position of the defendant in every case in which punitive damages are awarded, and as a precondition of such awards.(16) An inquiry of this sort could involve questions of great complexity (for example, in the case of corporate defendants) and discovery may involve substantial expense. Moreover, there is a risk of abuse by plaintiffs of rights to discovery, in order to oppress and to pressurise defendants.

5.8 We believe that the focus of the assessment of what is required in order to punish and deter the defendants outrageous conduct should, initially, be on the nature of that conduct, but that a defendant should have the opportunity to show that, in his or in her particular financial circumstances, an apparently and otherwise fair punishment would cause him or her undue hardship, if he or she had to discharge it. We therefore recommend that:

(26) the defendant should be allowed to show that he does not have the means, without being caused undue hardship, to discharge the punitive damages award which the court has decided to grant; where the defendant satisfies the court that this is so, the court must award a lower sum which it considers avoids that hardship. (Draft Bill, clause 6(2))

What this amounts to is a rebuttable presumption of ability to meet a punitive award without undue hardship, with the burden of rebuttal (obviously) falling on the defendant.

5.9 We would add, however, that we do not anticipate defendants seeking to rebut the presumption in very many cases: plaintiffs are unlikely to sue defendants who are obviously not able to satisfy an award made against them.

5.10 We consider that the definition of the means of the defendant should be left for the courts to flesh out, except to the extent of making one point clear. We recommend that:

(27) our draft Bill should provide that the defendants means include the fruits of any contract of insurance against the risk of liability to pay punitive damages. (Draft Bill, clause 6(4))

5.11 In our view, if a liability to pay punitive damages will be fully satisfied by sums paid under a contract of insurance, there should be no room for defendants (or their insurers) to argue that the award which would otherwise be appropriate punishment would cause them undue hardship, and so ought to be reduced. Of course, if a liability to pay punitive damages is only partially covered by a contract of insurance, then the defendant would have to show that paying the unsatisfied part (the total award less any sum payable by the insurers) will cause him or her undue hardship.

5.12 Where a court does reduce an award on the basis of the defendants incapacity to pay an otherwise appropriate sum, we consider that it would be desirable if the court not just found but also recorded the sum it would have awarded, but for the deduction. We therefore recommend that:

(28) where a court has decided to award punitive damages, it must indicate the amount which it is minded to award, irrespective of the defendants means; (Draft Bill, clause 6(1)); and if the court has reduced an award of punitive damages on account of undue hardship to the defendant (under recommendation (26)) the court should record what sum would have been awarded, but for that reduction. (Draft Bill, clause 6(3))

5.13 In our view, this is necessary to facilitate comparisons and so consistency between punitive damages awards. There is also a statutory precedent in the Law Reform (Contributory Negligence) Act 1945: section 1(2) requires a court to "find and record the total damages which would have been recoverable" if they had not been reduced because of the plaintiffs contributory negligence.

(3) The destination of punitive damages awards

(a) The problem of the destination of punitive damages awards

5.14 An objection regularly made to punitive damages awards is that they result in a plaintiff receiving a windfall benefit. One possible answer to this objection is to dismiss it as wholly misconceived. The civil law is not concerned only with compensation, as shown by restitutionary awards. Once plaintiffs have established that their rights have been infringed, they have established an entitlement to a range of remedies, which include, in certain circumstances, non-compensatory punitive damages. Once one accepts that civil punishment is legitimate (as we do) there is no necessary objection to the victim of a wrong keeping the punishment exacted.

5.15 But while we reject the compensation-only dogma, we recognise the force of the view that punitive damages are a hybrid. Although they are awarded in respect of an identified civil wrong against a private individual, they nonetheless include a significant (even primary) public element - that is, the public interest in the punishment and deterrence of outrageously wrongful conduct. On this approach, it is arguable that the windfall objection would be most appropriately met by making either all, or a percentage (say 33%), of any punitive award payable to the state or some other public fund.

5.16 Of these two, we do not find at all attractive the suggestion that all of the punitive award should be payable to the state (or other public fund). This would normally remove any incentive for a plaintiff to claim punitive damages and would therefore normally nullify the point in retaining punitive damages.(17) A plaintiff who stands to receive nothing from an award of punitive damages normally has no reason to claim them and, given the costs involved in establishing that they are merited, a clear financial disincentive to do so. If, as we believe, punitive damages play a valid role in ensuring that the civil law is properly upheld, it must follow that, even if one does not wish to go to the lengths of encouraging plaintiffs to sue who would otherwise not have sued, one should at least ensure that those who do sue for compensation (or restitution) are not discouraged from also seeking punitive damages.

5.17 Much more attractive, therefore, is the compromise position whereby the state or some other public fund would receive a percentage (say 33%) of any punitive damages award.(18) This would reduce the size of any windfall obtained by plaintiffs, whilst arguably preserving the financial incentive to claim punitive damages. Several American states have adopted this course of action. While neatly reflecting the hybrid nature of civil punishment, this compromise does carry with it some serious difficulties, which we must now consider.

(i) The cost of diversion to the state or public body

5.18 Applying our recommendation that awards be moderate in size, we fear that the costs of administering such a scheme might be disproportionate to the amounts being recovered. There are several possible responses to this fear.

5.19 The first response would involve confining state diversion to large awards. However, this would involve a cut-off point which might be vulnerable to abuse by more powerful parties and to accusations of arbitrariness.(19) A second and more general response would be to challenge whether the cost of diversion would be significant. For reasons stated below, only if the court makes an award in such an action, and the defendant fails to satisfy the part that is to be diverted to the public fund, will the significant costs of enforcement by the body in charge of the public fund arise.(20)

(ii) The enforcement of the punitive damages award

5.20 If the state or some other public body was to be entitled to a percentage of the punitive damages awarded, how would that part of the award be enforced? In a civil action, the victim of a wrong has the entitlement to sue and to bring further proceedings to enforce the award. However, he or she would have no incentive to bring further proceedings for the enforcement of any part of the award that is to go to the state. And certainly we would not favour a system whereby the claim of the victim-plaintiff is rendered secondary to that of the state or of a public fund. On the contrary, the claims of victim-plaintiffs should be the first to be satisfied from any sums received from defendants or their insurers.

5.21 However, while unusual, we see no objection to giving the state (or public fund) a right to enforce that part of the punitive damages award to which it is entitled. A possible analogy is with the role of the Attorney-General, who possesses the legal power to institute private law proceedings in his capacity as the guardian of the public interest. In our view, therefore, there would be no objection (other than cost, to which we have referred above), to conferring the right to enforce part of the punitive damages award on the Attorney-General (on behalf of the state) or on the trustees of the relevant public fund.(21)

(iii)Settlements

5.22 Perhaps the most serious difficulty with the 33% diversion suggestion would be the tactical incentives created by such a scheme. In order to catch all punitive awards it would be necessary for the scope of the scheme of diversion to include settlements. The primary difficulty is that where damages are settled by means of agreement between the parties, rather than by a court, the true nature of such damages can be distorted. In particular the incentive is likely to be for the parties to overstate the size of the compensatory (or restitutionary) damages award, so as to avoid diversion of any relevant part of the punitive damages award. On the other hand, one can argue that, as a counter-balance to the expansion of the scope of punitive damages, it is desirable to provide plaintiffs with an incentive to settle out-of-court.

5.23 Three further problems with awarding part or all of punitive damages to the state were put to us by consultees. We consider that each is plainly surmountable.

(i) "If the state takes part of a punitive damages award, it is imposing a fine"

5.24 Some consultees objected that for the state to take part of a punitive damages award is tantamount to imposing a fine at the instigation of an individual; this would constitute an unsatisfactory confusion of the criminal law and the civil law.

5.25 We are unconvinced by this. The first, essentially presentational concern, can be met by using the award in support of a valuable social cause. An often used justification for the availability of punitive damages is that they are necessary to enable plaintiffs fully to vindicate their rights which the defendant has infringed. It would be consistent with this, for example, to utilise any part of the award diverted by the state in the advice and even the financial assistance of other victims of civil wrongs. The second is to concede the analogy, but to deny its significance. All punitive damages, whether payable to the state or not, can be regarded as a type of fine.

(ii) "It infringes the rule that plaintiffs can do what they want with their damages"

5.26 It is a fundamental rule underlying the awarding of damages within the civil law that plaintiffs are entitled to use any damages they are awarded as they wish; this is infringed, it is argued, by requiring plaintiffs to pay all or any part of their award to a public fund.

5.27 However, this argument fails to recognise the distinctive, hybrid nature of a punitive damages award, and in any case is somewhat circular.(22) To assert that plaintiffs must be permitted to do what they like with their damages leaves unexplained when and why the damages are to be characterised as theirs, apart from the fact that they have sued to claim them. In the case of a compensatory damages award the explanation is clear and strong: the damages award is the plaintiffs because it represents the plaintiffs loss. It is, in a sense, an entirely private award. In contrast, in the case of a punitive award, which is inherently a non-compensatory award, the position is different. Here the award has a significant public element - corresponding to the function of the award in punishing and deterring the defendant and others minded to act in a similar way. It does not seem to us, therefore, that the normal rule for compensatory damages necessarily has to apply.

(iii)"It is unsatisfactory that the state should both pay and receive an award"

5.28 Some consultees observed that, where an organ of the state is made liable to pay punitive damages, one would be left with the unsavoury sight of the state being the beneficiary of the award which it had paid by way of punishment.

5.29 However, this argument could be dealt with, in two different ways. One response would be to exempt successful punitive damages claims against state organs from the category of awards where all or part of the punitive damages should be paid over to the state. To make this concession, however, would really render pointless - because even less cost-effective(23) - any form of diversion. A second response would be to use any diverted sum, not in some general way for the benefit of the state, but instead for the benefit of individuals in a similar position to the plaintiff.

5.30 In common with the views of a bare majority of consultees, we would recommend, albeit with some hesitation, that:

(29) no proportion of a plaintiffs punitive damages award should be diverted to a public fund.

Our main reason is that, since we anticipate punitive damages being moderate in size, we consider that the benefits of diversion would be outweighed by the costs involved and the tactical distortions in settlements that it might produce.

(4) Multiple plaintiffs

5.31 Multiple plaintiff cases raise very difficult practical problems, as well as problems of principle, for any framework of liability to punitive damages. Indeed, the fact that a case involves multiple claims to punitive damages, arising out of the same course of wrongdoing, has been considered a good reason for precluding any claim to that remedy.(24) Yet this means that some grave instances of wrongdoing - inter alia, in terms of the numbers of persons harmed - must go unpunished by the law. We are extremely reluctant to accept this result, unless compelled to do so by the clear absence of any practicable solution.

5.32 Very little assistance can be obtained from existing English and Commonwealth jurisdictions in resolving this issue. Nor does an awareness of the various approaches adopted in the USA,(25) where multiple plaintiff issues regularly arise, do other than reinforce the perception that this is an intensely difficult area.(26) We have therefore found it necessary to devise our own scheme.

5.33 One course of conduct may constitute or involve wrongs against more than one person; each victim may have a separate cause of action. Where the course of conduct is not just wrongful, but also punishment-worthy, then the apparent corollary is that each plaintiff should have a claim to punitive damages. In such circumstances there is a real risk that the defendant may be excessively punished.

(b) Our basic principle: first past the post takes all

5.34 We consider that the plaintiffs who are first past the post must take all. This has several implications. The first action in which punitive damages are awarded to one or more multiple plaintiffs will be the only action in which they can be awarded by a court (the first successful action). The defendants liability to pay punitive damages for the conduct that is punished in that action is thereafter extinguished;(27) thus no multiple plaintiff has any right to claim any further sum of punitive damages in respect of it. Furthermore, even if other multiple plaintiffs have well-founded claims to punitive damages, they will have no right to any part of the award(s) made in the first successful action.

5.35 We explain the scheme by which we propose to implement the first past the post takes all principle in the next section. We then defend what may, at first sight, appear to be an objectionable approach.

5.36 Although the core of our approach is aptly expressed by the principle that the first past the post takes all, the practical implementation of this approach raises several difficult questions. Who are multiple plaintiffs to whose claims to punitive damages we seek to apply the first past the post takes all principle? How should courts approach the issue of the availability of punitive damages if they are faced with claims by multiple plaintiffs? And, having decided that punitive damages are available to one or more multiple plaintiffs, how should those damages be assessed?

(i) The concept of a multiple plaintiff case

5.37 We consider that special provisions are necessary wherever conduct of a defendant constitutes torts, equitable wrongs or statutory wrongs against two or more persons.(28) The reason is that it is in this situation that the risk of excessive punishment, described above, is at its most severe. It is irrelevant that the wrongs which have been committed may be legally different; what is important is that certain conduct of the defendant may give rise to allegations by two or more people, in a single action or in a succession of separate actions, that a wrong has been committed against them. The potential for multiple claims to awards of exemplary damages in respect of the same conduct of the defendant is the same, whether that conduct is alleged to constitute, for example, the tort of trespass vis-à-vis A and B, or the tort of trespass vis-à-vis A and the tort of nuisance vis-à-vis B. We therefore recommend that:

(30) our special multiple plaintiffs scheme should apply where conduct of a defendant involves torts, equitable wrongs or statutory wrongs against two or more persons. (Draft Bill, clause 7(1))

(ii) The availability of punitive damages

5.38 In order to be entitled to an award of punitive damages, a multiple plaintiff will need to satisfy all the conditions which must be satisfied by an ordinary claimant to such a remedy.(29) However, in order to deal with the rather difficult problems that are raised by multiple plaintiff claims, we propose an additional limitation. The limitation which we recommend is as follows:

(31) once punitive damages have been awarded to one or more multiple plaintiffs in respect of the defendants conduct, no later claim to punitive damages shall be permitted for that conduct by any multiple plaintiff. (Draft Bill, clause 7(4))

5.39 This provision means that a multiple plaintiff will need to satisfy one additional condition if his or her claim is to succeed: there must have been no previous action brought by one or more other multiple plaintiffs in which punitive damages have been awarded in respect of the defendants conduct.

(iii)The assessment of punitive damages

5.40 We consider that a court should make a separate assessment of punitive damages for each multiple claimant. That is, the court should decide upon an appropriate sum by reference to the circumstances of the particular plaintiff before it. One plaintiff may have provoked the defendant to act in such a way that he or she committed a wrong against the provoking plaintiff and several others. If so, it is likely that the award (if any) which is made to the provoking plaintiff will be significantly less, in the light of his or her responsibility for the wrongful conduct, than any which is made to the other, non-provoking plaintiffs. Where the defendants conduct vis-à-vis the plaintiffs is essentially the same, it would obviously be open to a court to make identical individually-assessed awards to each of the plaintiffs.

5.41 Nevertheless, we do consider that a special approach must be taken to assessments of punitive damages in multiple plaintiff cases. It should be laid down in statute that the aggregate award of punitive damages to two or more multiple plaintiffs should conform to what we call the principles of moderation and proportionality, which apply to limit the assessment of individual punitive damages awards. In other words, the aggregate award should not punish the defendant excessively for his conduct. We therefore recommend that:

(32) if the court intends to award punitive damages to two or more multiple plaintiffs in the same proceedings, the aggregate amount awarded must be such that, while it may properly take account of the fact that the defendant has deliberately and outrageously disregarded the rights of more than one person, it does not punish the defendant excessively for his conduct. (Draft Bill, clause 7(3))

5.42 This express limitation on the total level of punitive damages awards in multiple plaintiff cases is, in effect, an application of the principles of moderation or proportionality (which are expressed in clause 5(2) of the draft Bill).(30) But for two reasons, we think that such a special statutory limitation is still required. The first is that our assessment provisions (in particular, clause 5 of the draft Bill) are otherwise directed only at individual assessments. As a result, the principles of proportionality and moderation (in clause 5(1)) prima facie only apply to require that the award which the court is making for the defendants conduct vis-à-vis a particular individual be proportionate and moderate. Without further provision they do not furnish a separate limitation - the requirement that the aggregate of a number of awards made in one action to a number of plaintiffs be moderate and proportionate (or not excessive). The second reason is that, for the reasons discussed above, the risk of excessive punishment is especially acute in multiple plaintiff cases. A special, express provision against excessive punishment should better direct courts to this risk.

5.43 What will this limitation require in practice? It may happen that the individually-assessed punitive damages awards, if they are added together, constitute excessive punishment for the defendants conduct. In order to avoid that excess, the court will obviously have to decide what an appropriate (lower) total liability is; it will then need to reduce each successful plaintiffs punitive damages award, so as to ensure that the aggregate is equal to that appropriate sum. We consider that a form of pro rata deduction from each individual punitive damages award would be the best solution. That deduction could proceed as follows:

Example:

A, B, & C are given punitive damages of £10,000, £10,000 and £20,000, in one action.

Applying clause 6(3) of our draft Bill, the court decides that the aggregate amount (£40,000) punishes the defendant excessively for his or her conduct; £30,000 would be sufficient. The aggregate amount is therefore £10,000 too much.

The awards are reduced by £10,000, preserving the proportion which they bore to the aggregate sum: A (1/4); B (1/4); C (1/2), or ratio 1(A) : 1(B) : 2(C). Accordingly, As award is reduced by £10,000/4 (£2,500); Bs award is reduced by £10,000/4 (£2,500) and Cs award is reduced by £10,000/2 (£5,000).

This leaves the final judgment as £30,000 in total, consisting of £7,500 (A); £7,500 (B); and £15,000 (C).

5.44 Where there are multiple plaintiffs or potential plaintiffs, the otherwise desirable practice of out of court settlement raises particular problems. A defendant may settle with some, but not all, potential multiple claimants. Unless there is at least the chance that this will be taken into account by a court, when deciding the defendants liability to punitive damages to plaintiffs who have not settled, the law could give a strong disincentive to defendants to seek to settle out of court, except where the defendant could be sure of securing a settlement with all potential claimants. This is because such a defendant will owe or have paid the settlement sum, but in addition will be liable to pay, inter alia, a sum of punitive damages which ignores the fact that he or she has settled with one, some or many potential claimants. The defendants total liability (settlement sums + court award) could be an excessively punitive sum.

5.45 In order to avoid this risk, we suggest that, in deciding whether punitive damages should be awarded and/or how much should be awarded in a multiple plaintiff case, the court should take account of any settlement which the defendant has made with other multiple plaintiffs in relation to the conduct. But this should only be so where the defendant consents to the court doing so. Were it otherwise, problems could arise because of the confidentiality of settlements. In practice, therefore, the onus would be on the defendant to bring any settlement to the attention of the court if he or she wishes it to be taken into account.

(33) provided the defendant consents to this, a court should take into account any settlement which the defendant may have reached with multiple plaintiffs in deciding:

(a) whether punitive damages are available, or

(b) if so, how much should be awarded

to multiple plaintiffs with whom the defendant has not reached a settlement. (Draft Bill, clause 7(2))

5.47 This proposal will mean that settlements may be a reason, depending on the circumstances, for a court to refuse punitive damages to multiple plaintiffs; or, in the alternative, for awarding a lower amount than would otherwise be appropriate (because, in particular, the aggregate of the settlement sums and the sums which the court is minded to award will excessively punish the defendant for his or her conduct).

(d) Some potential objections to our scheme and our response to those objections

5.48 The first past the post takes all principle may, at first sight, appear objectionable. We now review, and then respond to, likely objections.

(i) "It is unfair to deny punitive damages to multiple plaintiffs who are not parties to the first action in which a claim to punitive damages succeeds"

5.49 The first objection is that it is unfair to deny an award of punitive damages to a multiple plaintiff for the sole reason that one or more other multiple plaintiffs have already been awarded punitive damages in respect of the (same) conduct of the defendant. If they can establish an otherwise good claim, do they not have a right to an award of punitive damages, or if not, to some share in the awards that have previously been made? Were they not equally (or conceivably, more) wronged by the defendants conduct?

5.50 The second objection is to excessive windfalls. By restricting the entitlement of multiple plaintiffs to receive punitive damages to those of their number who are successful in the first action in which punitive damages are successfully claimed by multiple plaintiffs, first successful claimants may be left with very substantial awards of punitive damages. If, as we accept, an award of punitive damages is always a windfall to a plaintiff who receives it, does that vice not increase as the size of the award to individual plaintiffs increases?

(iii)"The first past the post takes all approach encourages multiple plaintiffs to race to court"

5.51 A third objection is that the first past the post takes all approach encourages a race to court. If the aggregate award made to successful multiple plaintiffs will not be substantially greater merely because the court has more multiple plaintiffs before it, then is there not a financial incentive for multiple plaintiffs to proceed alone, or in as small a group as possible? The fewer the people to whom punitive damages are awarded, the larger the likely entitlement of any particular successful individual to punitive damages.

5.52 A fourth objection is that the first award(s) bar and the principle of avoiding excessive punishment will not always be adequate to prevent defendants from being excessively punished in multiple plaintiff cases. Nothing in our proposals affects the entitlement of multiple plaintiffs to claim other remedies. Accordingly, even if a multiple plaintiff, who was not a party to the first action in which a punitive damages award was made, could not claim punitive damages in any later action, his or her claim to, inter alia, compensatory damages is not affected. Such subsequent claims arguably pose a risk of excessive punishment, because they could falsify the basis on which the award was made in the first action.

5.53 This problem relates to the if, but only if test. This will have to have been satisfied in the first successful action (because it is a pre-condition of any award of punitive damages). In other words, the court will have to have considered that the other remedies available to it were inadequate to punish and deter. The sum it awarded as punitive damages would reflect the extent of the inadequacy of the other remedies then available to it. But, of course, the defendants liability is not limited to a liability to those victims of his wrongdoing who are before the court in the first successful action; it also includes a liability to any other person who can subsequently show that the defendants conduct constituted a wrong against them - and in particular, a liability to pay compensation.(31) Accordingly, the defendants total liability for wrongs which he or she committed by one course of conduct may subsequently (that is, after the first successful action) be found to exceedthat which the court had assumed as a basis for deciding whether punitive damages were necessary in the first successful action.

5.54 We are not persuaded that any of these four objections fatally undermine the first past the post takes all approach. In particular, a plaintiff cannot assert as strong an entitlement to punitive damages as to compensatory damages, because punitive damages are always a windfall to plaintiffs who receive them. Indeed, we have seen that it is a controversial question whether any plaintiff should receive punitive damages (rather than, for example, the state); our justifications for plaintiff-receipt were practical, rather than doctrinal ones.(32) The first past the post takes all principle does not affect a plaintiffs right to other remedies.

5.55 We also think that any adverse effects of the first past the post takes all principle can be removed or diminished, if necessary. Underlying each of the above concerns is the assumption that, in practice, the bar will lead to punitive damages being awarded to only a very small proportion of potential (and potentially successful) claimants. We are not persuaded that this assumption is generally a correct one. Procedures for joinder or consolidation already exist which can be used by parties/courts to ensure that actions in which punitive damages are claimed by multiple plaintiffs include at least a substantial number of likely claimants.(33) Incentives to use those procedures may well be present. For example, a potential claimant has an incentive to join an action which has already been initiated by other victims of the defendants conduct, and which is likely to reach judgment before any action which the former subsequently initiates could do so. If that earlier action turns out to be the first successful action, any potential or actual claimant who was not a party to that action will be barred from claiming punitive damages. A court which is aware that two or more actions are in progress, arising out of one incident, may well be able (and willing) to consolidate the actions, on its own initiative or on an application - particularly because of similarity between the issues of fact and law raised, and because of the adverse effect of not being party to the one action in which punitive damages are awarded (that is, loss of the right to claim punitive damages). Indeed, in mass tort cases, the incentives for plaintiffs to join together, pooling information, resources and costs, may be sufficiently great that, even with the enticement of a large(r) award of punitive damages, a race to court is unlikely. This may a fortiori be the case, given the difficulty in such cases of establishing (at least) reckless wrongdoing, which is outrageous in character.

5.56 Even if practice reveals this belief to be misguided, we believe that it is a problem that can be dealt with, if and when it arises, without requiring any alteration to the first past the post takes all principle expressed in our statutory scheme. For example, the Rules Committee could develop procedures and powers for courts to deal with problems which are revealed in practice. These might include, for example, a notice-giving procedure, whereby a court, considering that there are multiple plaintiffs (present or potential) who are not parties to the action before it, could order that notice be given, in order to alert those others to the action before the court, and offer them an opportunity to obtain joinder or consolidation. Such a reform could be tied in with Lord Woolfs reforms (if and when implemented),(34) and with any general initiative on reforming procedure for multi-party actions generally.(35)

5.57 Some palliative could, if necessary, also be provided by the safety-valve discretion.(36) A court might, if it was aware that the plaintiff or group of plaintiffs before it represented only a small proportion of present (or perhaps also, likely) litigants, refuse to make any award in that action by exercising the last resort discretion. That a court had exercised the discretion to prevent an award going to the first individual or group of successful claimants in one case should not prevent an award being made to a subsequent (substantial) group of claimants in a later action. Indeed, such a later award would be entirely consistent with the justification for exercising the safety-valve discretion to preclude an award in the earlier action. If this became established judicial practice, then first plaintiffs would have a much reduced financial incentive to race to court to claim punitive damages.

(5) Multiple defendants

5.58 The law may regard the liability of two or more persons as either joint or joint and several in a number of different circumstances. For example, two or more people may independently act in a wrongful manner, and thereby cause the same indivisible damage to another. In law they are jointly and severally liable to compensate the plaintiff for that indivisible damage. Alternatively, two or more people may be regarded by the law as joint wrongdoers because they have taken concerted action to a common (wrongful) end. In law they are jointly liable to compensate the plaintiff for the injury which he or she suffers as a result of the joint wrong. Employers are vicariously liable for the torts of their employees (or servants). In law they are jointly liable with their wrongdoing employee to compensate the plaintiff for injury which he or she suffers as a result of those torts. Similarly, partners are jointly and severally liable by statute(37) for the wrongs of their co-partners. What should be the individual liabilities to pay punitive damages (if any) in each of the above cases, where not all of the wrongdoers acted in a punishable, or similarly punishable, manner?

5.59 For reasons which we explain below,(38) we consider that vicarious liability, as well as the liability of partners for the wrongs of their co-partners, should be treated somewhat differently from other examples of multiple defendants. Accordingly, the discussion which follows deals only with the other examples of multiple defendants.

(a) The problems of the existing approach

5.60 In principle, one would expect that punitive damages should be payable, but only payable, by those who have acted in a manner which warrants punishment, and only to the extent necessary to punish them for what they have done. The existing law in England does not reflect this principled conclusion.

5.61 The English common law attempts to fit liability to punitive damages within the framework of joint and joint and several liability (hereafter simply concurrent liability) which applies to any compensatory damages to which joint or joint and several tortfeasors (hereafter simply concurrent tortfeasors) may be liable.(39) But this is a very poor fit. What works satisfactorily for compensatory damages produces anomalous results when applied to non-compensatory damages.

5.62 In Broome v Cassell(40) it was held that where a plaintiff brings a single proceeding against two or more concurrent tortfeasors:

(1) only one sum of punitive damages can be awarded against the joint defendants; but

(2) that sum must not exceed the sum which is necessary to punish the least culpable of the joint defendants.

5.63 The consequences are objectionable. The limitation represented by proposition (2) is clearly required in order to avoid the risk of over-punishment or even of punishment where none is justified. Without it a concurrent tortfeasor might be made liable to pay an award which was available because of, and/or was assessed with reference to, the greater fault of another concurrent tortfeasor. If so, the punitive award would inevitably exceed that which was proportional to his or her fault, and necessary to punish him or her for it. And if the other concurrent tortfeasors were insolvent, so that contribution would not be possible, he or she would have to bear the full burden of that inappropriate award. However, by avoiding the risk of over-punishment by proposition (2), the present approach may leave concurrent tortfeasors under-punished or even unpunished. Where two or more of such tortfeasors are jointly sued, the liability of any will at most be that of the least culpable of their number;(41) the logical corollary of this principle is that if no award is warranted by the conduct of that tortfeasor, the maximum liability of each will be nil.(42) Highly culpable tortfeasors are fortuitously benefited for no better reason than that they happen to be associated with less culpable tortfeasors.

(b) The preferable, principled approach: several liability

5.64 A number of Commonwealth courts have refused to follow the English approach. Australian and Canadian courts recognise (in effect) several liability to punitive damages.(43) Separate awards of punitive damages, for different amounts, may be made against each individual joint tortfeasor. This means that if only one joint tortfeasors conduct justifies an award of punitive damages, judgment for such damages will be entered against that wrongdoer only, and the sum awarded will be that which is appropriate to his or her conduct alone. It also means that, if an award of punitive damages is justified by the conduct of two or more joint tortfeasors, separate judgments for punitive damages will be entered against each of them, for such sums as are warranted by their personal conduct. The position adopted by these courts has wide support from inter alia, Commonwealth academics(44) and authorities,(45) and the Ontario Law Reform Commission.(46)'

5.65 'We consider that the only principled and workable way forward is to follow the example of other Commonwealth jurisdictions, and introduce several liability to punitive damages.(47)' Our provisional view to this effect(48)' was supported by the overwhelming majority of consultees. A wrongdoer should be liable to a punitive damages award only where such award is available because of, and is assessed with reference to, his or her personal conduct. Separate punitive awards would be made against each individual wrongdoer; there will be no prospect of contribution.(49)' This is the only way in which it can be ensured that a wrongdoer is made liable to pay a sum which is simultaneously effective, but not excessive, and which is moderate and proportionate to the gravity of his or her wrongdoing. We therefore recommend that:

''(34) several liability, rather than joint or joint and several liability, should apply to punitive damages (subject to recommendation (35) below); '(Draft Bill, clause 8(1))

5.66 ''The implications of our recommendation that the liability to punitive damages should be several, rather than joint or joint and several, can be illustrated by two examples.

''Two concurrent tortfeasors cause (the same) damage to the plaintiff, but only one of them acts outrageously in the sense required before a punitive award is available to a court.'

''In this case, both of the tortfeasors will be jointly and severally liable to pay the compensatory award; however, only the one (highly culpable) defendant will be liable to pay the punitive award - in full and without the prospect of contribution.

''Two concurrent tortfeasors cause (the same) damage to the plaintiff, and both of them act outrageously in the sense required before a punitive award is available to a court.'

''In this case, both tortfeasors will be jointly and severally liable to pay the compensatory award; each could also be liable to pay separate punitive awards, assessed by reference to the personal conduct of each - which will be payable in full and without the prospect of contribution. It will be open to a court to decide that one of the outrageous tortfeasors should not be punished by a punitive damages award. This might be because, for example, he or she has already been convicted in criminal proceedings for an offence which involves the conduct alleged to be outrageous, whereas the other tortfeasor has not been subjected to criminal proceedings. Or it might be because, for example, the court intends to award a remedy to the plaintiff against him or her, but not against the other tortfeasor (such as a substantial award of restitutionary damages), which is adequate to punish him or her for his or her conduct.

5.69 ''Whilst we consider that several liability is the most justifiable general approach to cases involving multiple defendants, several complications arise:

5.70 ''If several liability is adopted, the windfall-to-plaintiffs objection which is raised against punitive damages awards may appear to apply with substantially greater force. Several liability has the effect that each defendant may be made liable to pay a separate sum which is made necessary by his or her personal conduct alone. The potentially problematic implication is that the plaintiff in such an action could receive as many punitive awards as there are defendants.

5.71 'On reflection, however, we do not consider this to be a decisive objection to several liability for punitive damages awards. Take the following example:

''A person (P) is assaulted by two people (D1 & D2), acting independently of one another. In case A, D1 and D2 both choose to kick P in the abdominal region, cumulatively breaking Ps ribs. In case B, D1 kicks P in the same region, breaking Ps ribs, but D2 kicks P in the head, putting him into a coma.

''In case A, D1 and D2 are joint and several tortfeasors; in case B, they are not. No objection can surely be raised to two punitive awards being made and received in case B, but on the view expressed above,(50)' objections will be raised to the plaintiff receiving two awards in case A. Yet the only factual distinction between cases A and B is an immaterial one as far as punishment (as opposed to compensation) is concerned: namely, that D1 and D2 have independently acted to cause the same' (case A) or different' damage (case B) to the plaintiff. In cases A and B alike, two very grave wrongdoers require punishment for the deliberate and outrageous disregard of the plaintiffs rights which each has respectively shown.

5.72 ''The absence of any sound basis for the initial objection that several liability will leave a plaintiff with an unjustified windfall can be further appreciated if one compares the scenarios outlined above with a different situation, in which it will be uncontroversial that two awards should be made and received by the plaintiff: that is, where the independent acts of D1 and D2 occur a significant time apart from each other.

5.73 ''The last resort nature of the punitive award (whereby an award is permitted only where any other remedy or remedies which the judge is minded to award will be insufficient to punish and deter) means, in particular, that the size of a punitive award will vary as the size of the compensatory award varies. Thus, if the compensatory award is small, a higher punitive award may be necessary in order effectively to punish and deter.

5.74 ''The source of the complication here is that joint or joint and several liability must continue to operate in respect of the compensatory part of any award. As a result, the size of such award could fluctuate substantially, depending upon whether the defendant was able, or unable, to obtain contribution from the other wrongdoers. This poses a risk of over or under-punishment of defendants. The problem is that the court, in deciding whether to make an award, and if so, at what level, must inevitably make some assumption about the defendants chances of obtaining contribution. If the court makes an award on the basis of an erroneous assumption that the defendant will get contribution, then the defendant will be punished and deterred to a greater extent than the court thought to be both necessary and proportional to the outrageousness of his conduct. For having underestimated the ultimate size of the compensatory award, the court will have assessed the punitive award at too high a level.

5.75 ''However, we anticipate that, if there was any doubt about the matter, a court would assess a punitive award on the basis that the defendant will be liable for the whole of the compensatory award (that is, irrespective of the availability of contribution). Under-punishment and under-deterrence are less undesirable than leaving a defendant over-punished and over-deterred. The core principles of moderation and of proportionality, which are each a vital part of the laws protection of the rights of wrongdoers, require this.

''(iii) Exceptions to the general principle of several liability to punitive damages: vicarious liability and partnerships

5.76 ''We consider that two exceptions are required to the principle that liability to punitive damages should be several, rather than joint or joint and several. The first is the doctrine of vicarious liability. The second is the liability of partners for wrongs of their co-partners. We therefore recommend that:

''(35) recommendation (34) (several liability, rather than joint or joint and several liability shall apply to punitive damages) is without prejudice to:

''(a)our recommendation that vicarious liability to pay punitive damages should be retained; '(Draft Bill, clause 8(2)(a))

''(b)the liability of a partner for the wrongs of his co-partner. '(Draft Bill, clause 8(2)(b))

''We explain the exception for vicarious liability below,(51)' but it is convenient at this point to explain the exception for the liability of partners for co-partners.

5.77 ''Partners are jointly and severally liable to any persons who are not themselves partners for the wrongs committed by any partner acting in the ordinary course of the business of the firm or with the authority of his co-partners.(52)' That liability is expressed to include a liability for "penalties" imposed as a result of the wrongful conduct.(53)' Prima facie' section 10 of the Partnership Act 1890 also makes partners jointly and severally liable to pay punitive damages in respect of the wrong of a co-partner. We have been unable to discover any case in which partners have been held to be so liable, or in which the point is even discussed. But we do consider that this could properly occur. Accordingly, our proposal that any liability to punitive damages should be several (rather than joint, or joint and several) is subject to the qualification that it should not affect the (joint and several) liability of innocent partners to pay punitive damages in respect of the wrongs of a co-partner.

''(iv) The right to contribution under the Civil Liability (Contribution) Act 1978

5.78 ''The Civil Liability (Contribution) Act 1978 provides that any person who is "liable in respect of any damage" suffered by another may recover contribution from any other person who is "liable in respect of the same damage".(54)' A person is "liable in respect of any damage" if the person who suffered the damage is entitled to recover compensation from him or her in respect of it, whatever the legal basis of the liability.(55)'

5.79 'In general, it would be inappropriate for the statutory right to contribution to operate in respect of a liability to punitive damages. We recommend above that,(56)' subject to two specific exceptions, the liability to pay punitive damages should be several only.(57)' Where this is the case, there should be no right to contribution under the 1978 Act. For the avoidance of doubt,(58)' we accordingly recommend that:

''(36) our draft Bill should ensure that the right to recover contribution laid down in section 1 of the Civil Liability (Contribution) Act shall not extend to a liability to pay punitive damages that is several. '(Draft Bill, clause 8(3))

5.80 ''Where a liability to punitive damages is other 'than several, however, any right to claim contribution conferred by the 1978 Act should continue to exist. This means that where a person is held vicariously liable to pay punitive damages, or is a partner who is held jointly and severally liable to pay punitive damages in respect of the wrongs of a co-partner,(59)' he or she should not be prevented from claiming contribution from his or her employee, agent or co-partner, when the 1978 Act currently entitles him or her to do so.

5.81 ''The questions we address here are whether, and if so, when, a person should be held vicariously liable to pay punitive damages in respect of anothers wrongful conduct.(60)' Although it has consistently been assumed that vicarious liability extends to exemplary or punitive damages on the same basis as compensatory damages,(61)' we cannot find any case in which the application of vicarious liability has been challenged in an English court. Existing authorities therefore offer little assistance in resolving this difficult issue. Several strong objections can in fact be raised to recognising vicarious liability to punitive damages.

5.82 ''The first objection is that vicarious liability imposes a burden on employers that is unfair, because it imposes the cost of an award on an innocent employer, whilst leaving the guilty employee unpunished by law. Prima facie' this is objectionable for precisely the same reasons as require that several liability to punitive damages be introduced in relation to joint and several tortfeasors. As the award will not be made against the primary wrongdoer-employee,(62)' it will not be effective; and as the award will be exacted from an innocent employer, it will necessarily infringe the principles of moderation and proportionality which we consider constitute vital limiting principles on the scope of liability to punitive damages.

5.83 'The second objection is that recognising vicarious liability for punitive damages imposes a burden on employers that is not warranted by the policies which serve to justify the laws recognition of claims to punitive awards. A similar objection has been raised against insurance against such awards.(63)' The immediate practical effect of insurance and of vicarious liability is that the burden of liability is transferred from the primary wrongdoer to another party - whether the employer or the insurer. As a result, the direct punitive, deterrent or symbolic efficacy of the punitive award is at best substantially diluted. The primary wrongdoer does not feel the punitive award in his or her pocket.

5.84 'We acknowledge the force of these arguments. Together they appear to entail that one should refuse to recognise vicarious liability to punitive damages. But for the reasons which we elaborate below, and in agreement with the majority of consultees, we nevertheless consider it to be correct to recommend that:

''(37) our draft Bill should clarify that a person may be vicariously liable to pay punitive damages in respect of anothers conduct;' (Draft Bill, clause 11(1))

5.85 ''Our reasons for preferring to recognise vicarious liability, which we elaborate fully below, can be summarised as follows:

(1) ''Vicarious liability arguably produces unfairness and gives rise to problems of justification even in relation to compensatory damages.

(2) 'In practice the objection that, by permitting vicarious liability, punitive damages are rendered ineffective, pointless and so unjustified, is not as strong as it may initially appear. Vicarious liability in respect of punitive damages may further the aims of such damages. And in some cases such liability may be the only way of furthering those aims.

(3) 'The situation of joint and several tortfeasors is materially different from that of employer-employee joint tortfeasors, so that the unfairness which is clear in joint and several liability to punitive damages is less clearly present in cases of vicarious liability.

5.87 ''It is notoriously difficult to find a convincing, comprehensive justification of the doctrine of vicarious liability.(64)' As a result, the unfairness objection stated above is not unique to vicarious liability to awards of punitive damages - a rather similar objection could be raised even against vicarious liability to compensatory damages. As P S Atiyah observes:

''Vicarious liability is one of the most firmly established legal principles throughout the common law world, but generations of lawyers have felt in some uneasy way that there is something so odd or exceptional about vicarious liability that it needs justification; and they have been hard put to it to justify the doctrine though almost unanimous in admitting that it is a laudable and necessary part of the law of torts.(65)'

''The reasons for this unease are probably two-fold. Vicarious liability seems at first sight to run counter to two principles of the law of torts, namely that a person should only be liable for loss or damage caused by his own acts or omissions, and secondly that a person should only be liable where he has been at fault. These principles are so deeply rooted in legal thinking that any departure from them seems at first sight impossibly unjust.(66)'

5.88 ''The objection of unfairness to vicarious liability to punitive damages is that innocent employers are punished for the wrongful acts of their employees. Yet as Atiyah recognises, vicarious liability to compensatory awards may also entail that an employer can be held liable even in circumstances in which he or she has not been at fault. Accordingly, there is one central similarity: a defendant-employer is liable to pay a sum of money which is not required to rectify any wrongful loss which he or she has caused or any wrongful gain which he or she has made, nor to punish him or her for any wrongful conduct of his or her own'.

5.89 ''In relation to vicarious liability in respect of compensatory damages, many would consider that the law rightly tolerates any perceived unfairness at least partly because, by allowing plaintiffs to proceed against solvent employers, their chances of having any claim to compensation satisfied are significantly improved. As a result, the primary compensatory aim of such damages is furthered, rather than frustrated, by the recognition of vicarious liability.

5.90 'On careful consideration we believe that, although contrary to what one might at first think, the same reasoning applies to awards of punitive damages. In other words, the primary punitive aim of such damages can be furthered, rather than frustrated, by the recognition of vicarious liability. If so, then this provides at least some reason why any perceived unfairness in vicarious liability ought to be tolerated. We explain this point below.

''(ii) Furthering the purposes of punitive damages by means of vicarious liability

5.91 ''The immediate impact of vicarious liability is clearly that the primary wrongdoer escapes punishment by the law'.(67)' Nevertheless, vicarious liability may offer a wider, if indirect, method for pursuing the aims of punitive damages.(68)'

5.92 'Employers who are so liable, or who are potentially so liable, will have an incentive to control and educate their workforces. The development by employers of some form of wrong-preventing educative process might be particularly beneficial, for example, in cases of sex or race discrimination. Employers also possess a range of disciplinary powers which will enable them to penalise and deter individual guilty employees, or to discourage potential wrongdoers. Indeed, the loss of employment, coupled with impaired employment prospects,(69)' may be a more severe form of sanction for wrongdoing by employees than a punitive damages award could directly provide.

5.93 'In two categories of case, moreover, vicarious liability may provide the only method for pursuing the aims of punitive damages. These are, firstly, where employees are unlikely to be able to satisfy a punitive damages award of any significant size; and secondly, where a plaintiff has problems identifying the culpable member of the employers workforce.(70)'

''(iii) Joint and several liability is distinguishable, and more clearly unfair

5.94 ''So far we have assumed that vicarious liability is prima facie unfair to employers; we have then suggested that this unfairness could be tolerated to the extent that, by recognising vicarious liability to punitive damages, the aims of punitive damages are furthered, rather than frustrated. 'We now seek to deny that there is, on closer examination, anything intolerably unfair about vicarious liability in respect of punitive damages. Our argument is that joint and several liability and vicarious liability are distinguishable in two important ways which indicate that the unfairness that we have identified in the former does not so clearly exist in the latter case.

5.95 'The first difference is that an employee forms part of the employers enterprise and that that enterprise generally profits from his or her employment. In most cases there will be no similar relationship between joint and several tortfeasors. This element of benefit to the employer is one justification which has been offered for vicarious liability to compensatory damages. This justification is founded on a moral imperative that one who derives a benefit - in particular a financial profit - from certain acts, should also bear the risk of loss therefrom. Jane Stapleton has recently offered a more sophisticated formulation of this argument, which she terms "moral enterprise liability".(71)' Such an argument does not appear to be any less applicable to vicarious liability in respect of punitive damages than to vicarious liability in respect of compensatory damages. Moreover, as Jane Stapleton observes, this argument does successfully explain key features of the doctrine of vicarious liability within the civil law. In particular, it explains the restriction of the liability of employers to liability for the acts of their employees within the course of their employment. Current tests used to determine whether a person is an employee or an independent contractor have at their core the question "who has the chance of gain and bears the risk of loss?" - or in a slightly different form, "is he or she in business on his or her own account?"(72)' The course of employment criterion also generally restricts the scope of liability of employers to those cases in which the employee is acting for the employer.

5.96 'The second difference is that, in very many cases, joint and several tortfeasors have neither the moral nor legal responsibility (or power) to control or influence the past and future behaviour of fellow joint and several tortfeasors. Nor, in most cases, will they have any practical ability to do so. Making a joint and several tortfeasor liable to sums awarded in respect of the conduct of others will therefore do little to advance the aims of punitive damages. The position of employers is precisely the reverse - as we indicate in argument (ii) above. Moreover, the control test was traditionally regarded as the 'criterion by which the legal status of employee was to be distinguished, thereby defining the category of persons for whose acts employers can be vicariously liable. And even today, when the control test has been replaced by a multiple factor approach, control remains one such important and relevant factor.(73)'

5.97 ''When the doctrine of vicarious liability is discussed in cases and texts on the law of tort, those discussions are first and foremost about a persons vicarious liability to pay compensatory damages. Nevertheless, we, consider that if, as we have argued above, vicarious liability to punitive damages can be justified, it should apply subject to the same conditions as apply generally(74)' to the wrong in question. Several considerations support this view. First, to apply two different concepts to different parts of the same claim(75)' may promote excessive complexity in argument and adjudication. In particular, it could produce extensive debate as to the precise distinctions between the ordinary and the new (and narrower) concept. Secondly, this complexity would be unwarranted because, we believe, the concept of vicarious liability, as generally formulated, adequately defines the category of case within which it is fair (or at least not unfair) to make employers liable for the wrongs of their employees.(76)' Thirdly, it would be unfortunate if our Act made a person vicariously liable to pay punitive damages for a wrong committed by another if the concept of vicarious liability has never been recognised in relation to (or deliberately excluded from) the wrong in question.

5.98 ''At common law, employers are vicariously liable to pay compensatory damages for the torts of their employees, committed within the course of their employment.(77)' In our view, the same concept should generally define when an employer is liable to pay punitive damages for its employees torts. However, statutes occasionally expressly formulate the doctrine of vicarious liability for the purposes of a particular wrong (such as unlawful discrimination on grounds of sex, race or disability),(78)' or extend the doctrine to apply to persons who are not, strictly, the employers of the employee in question.(79)' In either case, it is the concept so formulated which should define the extent of vicarious liability to pay punitive damages. And if a certain person,(80)' or persons generally,(81)' can never be vicariously liable for a particular wrong (apart from our Bill), our Bill should not make such a person or persons liable to pay punitive damages for that wrong.

''(38) our draft Bill should not define the circumstances in which one person may be vicariously liable for the wrongs of another; instead, it should assume the boundaries of the concept of vicarious liability as it exists at common law, or by statute, for the particular tort, equitable wrong or statutory wrong in question.' (Draft Bill, clause 11(1) and 11(2))

5.100 ''What sum of damages should a person, who is vicariously liable for the wrong of another, be liable to pay? The nature of vicarious liability should generally entail that the sum should be that which that other is or would be liable to pay. Thus, if faced with an employer who (it is alleged) is vicariously liable for the wrong of his or her employee, the court should determine what punitive damages the employer is liable to pay by applying the tests of availability and the principles of assessment to the conduct of the employee' for whom the employer is vicariously liable.(82)'

5.101 'There is, however, one important reason why a person who is vicariously liable to pay punitive damages for the wrongs of another may have to pay a different sum from that which the other is or would be liable to pay. We have recommended that defendants should be permitted to argue that they will suffer undue hardship if they must satisfy the award of punitive damages which the court proposes to make against them,(83)' and that, if this argument is accepted by the court, a lower award must be made. On this basis, employee-defendants will be liable to pay a reduced award if they succeed in persuading the court that they have insufficient means to satisfy the proposed liability. In our view, vicariously liable defendants should also be permitted to advance the argument that they (the employers) have insufficient means, but not the different argument that what they must pay should be reduced on account of their employees insufficient means.(84)' On this limited basis it is quite conceivable that employers could be liable to pay sums of punitive damages in excess of those which their employees are or would be liable to pay: primarily because an employee is much more likely than his or her employer to succeed with an argument that the proposed award is too high for him or her to pay.

''(39) subject only to recommendation (40), the sum of punitive damages which a person is vicariously liable to pay for the wrong of another should be that which that other would be liable to pay, and should be determined on that basis.' (Draft Bill, clause 11(2))

''(40) where the court is assessing the sum of punitive damages which an employer is vicariously liable to pay for the wrongs of its employee:'

''(a)the award payable by the employer may be reduced (in accordance with recommendations (26)-(28)) if the court considers that the employers means are such that it would cause it undue hardship to be required to pay such sum as would otherwise be appropriate, '(Draft Bill, clause 11(3)) and

''(b)the award payable by the employer must not be reduced on the ground that the employees means are such that it would cause the employee undue hardship if he or she was to be required to pay such sum as would (disregarding the means of the employee) otherwise be appropriate. ' (Draft Bill, clause 11(3))

5.103 ''We are content, in agreement with the majority of consultees, to continue to apply the civil standard of proof to claims to punitive damages.(85)' That this is the existing legal position was recently confirmed by the Court of Appeal in John v MGN Ltd'.(86)' We therefore reject any view that the criminal standard of proof is appropriate, in order to replicate the evidential safeguards that are offered by the criminal law, and in recognition of the quasi-criminal nature of the activity which may give rise to a punitive damages award. We also reject an intermediate standard, such as clear and convincing evidence, which has been adopted in some American states. Accordingly, we recommend that:

''(41) if it is sought to establish a matter relating to the question whether punitive damages should be awarded, or to the question of their amount, the civil, and not the criminal, standard of proof must be satisfied. '(Draft Bill, clause 10)

5.104 ''One reason for accepting the civil standard of proof is that that standard is, in fact, an inherently flexible standard.(87)' Clearer evidence will be required by the courts, in order for such standard to be satisfied, where the allegations, or the consequences of the decision for one or both of the parties, are serious. Both of these conditions will generally be satisfied by claims to punitive damages. The corollary is that defendants to such claims may be adequately protected even without the criminal standard of proof and within the lower civil 'standard. Cross & Tapper on Evidence' deals with the analogous case of allegations of criminal conduct in civil actions in the following way:(88)'

''... the person against whom criminal conduct is alleged is adequately protected by the consideration that the antecedent improbability of his guilt is a part of the whole range of circumstances which have to be weighed in the scale when deciding as to the balance of probabilities.

5.105 ''A second reason for accepting the civil standard is that it could be impractical for a higher burden of proof to be adopted for only one part of a civil action: all other aspects of liability (especially, for example, the commission of the wrong which founds a claim to compensatory damages) would be determined according to the ordinary civil standard; whereas just one aspect (that is, deliberate and outrageous disregard of the plaintiffs rights) would be determined according to the criminal standard.

5.106 ''In this section we mainly deal with the question whether a person should be permitted' to insure against any liability to punitive damages which they may incur (personal or vicarious). In the final subsection,(89)' we also deal with the rather different question of whether statutes which currently make insurance compulsory' in certain circumstances require, or should be construed to require, insurance against a liability to pay punitive damages (and not just compensation) for a wrong.

5.107 ''We have found the issue of whether a person should be permitted to insure against a liability to punitive damages difficult to resolve, not least because consultees put forward a very wide range of opinions. A survey of the approach of other common law jurisdictions to this issue similarly reveals a considerable range of approaches.(90)' Although it was not expressed in this way in Consultation Paper No 132,(91)' we regard the choice as one between three main options:

(1) ''insurance against punitive damages awards is in all cases permitted by legislation;

(2) 'there is no general legislative public policy bar on insurance, but insurance is barred in cases involving especially outrageous conduct;

5.108 ''On the balance of arguments of principle, policy and practicality, we reject a bar of any sort on insurance against punitive damages: that is, we favour option 1. We give the decisive reasons for our choice below.

''(b)The decisive reasons for preferring option 1: insurance is permitted in all cases

''(i)The need for plaintiffs to have a financial reason for claiming punitive damages

5.109 ''There is a clear public interest in punishing and deterring bad conduct of a nature which merits a punitive damages award, as well as in offering appeasement to the victims thereof. Nevertheless, it is futile to discuss the pursuit of these aims through civil litigation if plaintiffs will not claim punitive damages because the defendant cannot pay them. Plaintiffs are unlikely to claim punitive damages where defendants do not have the financial capacity to pay any substantial damages and costs which may be awarded against them. Such capacity may be afforded, however, by liability insurance.

5.110 ''We do not believe that the aims of punitive damages will be either wholly or substantially frustrated by generally permitting insurance against awards. Although we recognise that any retributive and deterrent purposes of this category of damages may be diluted by our proposed approach to insurance, we do not anticipate that they will be wholly frustrated: in particular, the insurance industry, in controlling the availability and cost of such insurance, is in a position to exert significant pressure on present or potential insured parties.

5.111 'Our views on this matter are supported by strong recent judicial statements. In Lamb v Cotogno'(92)' the High Court of Australia recognised that the purposes of punitive damages are not wholly frustrated by the availability of insurance:

''The object, or at least the effect, of exemplary damages is not wholly punishment and the deterrence which is intended extends beyond the actual wrongdoer and the exact nature of his wrongdoing.(93)'

5.112 ''In the more recent case of Lancashire County Council v Municipal Mutual Insurance Ltd',(94)' Simon Brown LJ in the Court of Appeal considered that, owing to the responses of the insurance industry, an exemplary damages award was "still likely to have a punitive effect":

''First, there may well be limits of liability and deductibles under the policy. Second, the insured is likely to have to pay higher premiums in future and may well, indeed, have difficulty in obtaining renewal insurance.(95)'

5.113 ''Moreover, regardless of the impact of permitting insurance against the possibility of awards of punitive damages on the aims of punishment and deterrence, the aim of satisfaction of the plaintiff can still coherently be pursued even where a defendant is insured. As a significant number of cases in this area could involve the violation of important rights of plaintiffs, yet no or very little compensatable loss, the importance of this aim ought not to be underestimated.

5.114 ''There is a general policy underlying the law of contract that commercial contracts ought not to be lightly interfered with by courts or even legislation. In Printing & Numerical Registering Co v Sampson',(96)' for example, Sir George Jessel MR offered a powerful entreaty to courts considering the application of any doctrine of public policy:

''... if there is one thing which more than another public policy requires it is that men of full age and competent understanding shall have the utmost liberty of contracting, and that their contracts when entered into freely and voluntarily shall be held sacred and shall be enforced by Courts of justice. Therefore, you have this paramount public policy to consider - that you are not lightly to interfere with this freedom of contract.(97)'

5.115 ''Courts have required that the reasons for imposing a public policy be forceful, and not open to doubt, before they will apply or extend a public policy bar. As Simon Brown LJ stated in Lancashire County Council v Municipal Mutual Insurance Ltd'(98)' in indicating his opposition to a public policy bar on insurance against awards of exemplary damages:

''... [c]ontracts should only be held unenforceable on public policy grounds in very plain cases.(99)'

5.116 ''This observation is particularly apt given that, as the diversity of responses which we received demonstrated, the case of insurance against punitive damages is by no means a very plain one.

5.117 'Such judicial caution has been demonstrated in relation to both the interpretation and extension of any common law bar and' the construction of statutes which may have the effect of rendering a contract illegal and so potentially unenforceable.(100)' Thus in St John Shipping Corporation v Joseph Rank Ltd'(101)' Devlin J dealt with the correct approach to statutory construction where in the performance of a contract statutory provisions have been breached, and it is alleged that the contract is (by that statute) impliedly rendered illegal and unenforceable:

''[Without a clear implication of statutory intention, courts should be] very slow to hold that a statute intends to interfere with the rights and remedies given by the ordinary law of contract.(102)'

5.118 ''At a more specific level we would argue that if insurers accept a premium to cover a certain risk, they should meet it. This point was also made by Simon Brown LJ in Lancashire County Council v Municipal Mutual Insurance Ltd'.(103)'

5.119 ''We were also impressed by two arguments put forward by consultees to the effect that a bar on insurance would be inconsistent or unfair in its effects, owing to some alternative ways in which a liability to pay punitive damages could be met without the need to insure, and even 'if insurance against such liability was to be barred. The first argument was that a bar on insurance is objectionable because it produces inequality between the impact of punitive damages awards on organisations which are able to self-insure and those which cannot. The second was that no objection is made to allowing another person or organisation to meet a defendants liability to punitive damages by way of a gift.

5.120 ''A final point is that to permit insurance against punitive damages may minimise the number of occasions on which defendants and their insurers' come into conflict - in the settlement process, or in court. Such conflict might arise in a case where the defendant is insured against the non-punitive part of any award, but did not or could not obtain insurance against the punitive award. In this situation, the defendant may seek to maximise the size of any compensatory or restitutionary award. This would serve to maximise the part of a total award that is covered by insurance. In direct conflict with the interests of the defendant on this issue, however, are the interests of the insurers. This is because their interests would lie in minimising the size of any non-punitive award, the risk of which they must meet, and in maximising the size of any punitive award, the risk of which they need not meet. Undesirable consequences, in particular the need for three sets of legal representation (for defendant, plaintiff and insurers), might ensue.(104)'

''(c)The reasons for rejecting option 2: a public policy bar in the case of particularly outrageous conduct

5.121 ''Option 2 was raised in one possible form before the Court of Appeal in Lancashire County Council v Municipal Mutual Insurance Ltd'.(105)' The defendants counsel argued that there should be a bar on insurance where the conduct which gave rise to the award of exemplary damages was criminal in nature. The Court of Appeal left open what should be the proper approach to cases involving the personal liability of defendants, but it rejected an option 2 approach, and adopted an option 1 approach, in relation to vicarious liability.(106)'

5.122 'We consider, however, that option 1, and not option 2, is the correct approach to adopt in relation to both personal and vicarious liability. This is so whether the appropriate characterisation of the cases in which a public policy bar applies is, for example, especially outrageous conduct or conduct constituting a criminal offence. In addition to the five positive reasons given above for favouring option 1, we consider that there are three specific reasons for rejecting option 2.

''(i)The greater need to preserve a financial reason for plaintiffs to claim punitive damages in the case of particularly outrageous conduct

5.123 ''First, and most importantly, the need for plaintiffs to have a financial reason to claim punitive damages has even greater force in the case of the particularly outrageous conduct which would be made the subject of a bar on insurance under option 2. Perversely, a bar on insurance in the case of particularly outrageous conduct would reduce, rather than increase, the prospect of punitive damages being claimed. This would not be in the public interest of securing the punishment of serious wrongdoers.

''(ii) The problems of defining with certainty the range of conduct falling within the public policy bar on insurance

5.124 ''Any concept which is used to define the category of particularly serious conduct which not only warrants a punitive damages award, but also justifies the further step of a public policy bar on insurance, should be capable of precise definition: legal and commercial certainty so require. We do not consider that a concept such as particularly outrageous conduct satisfies this requirement.

5.125 'A possible response is to adopt instead a concept which draws a parallel with conduct meriting prosecution within the criminal law. One example is conduct constituting a criminal offence. However, although this offers greater conceptual clarity, we consider it to be objectionable in principle.

5.126 'Our objections are three-fold. The first is that the judgment concerning the criminality of the defendants conduct would have to be made within a civil court; it cannot be assumed that such court would have the experience in dealing with such matters. The second, and more important, is that to utilise any such concept would involve denying a defendant the procedural and evidential safeguards found within a criminal trial; it is also open to accusations that any adverse or favourable finding could prejudice any subsequent criminal prosecution that may be brought. The third is that conduct constituting a criminal offence will not in fact capture, and capture only, the most outrageous examples of conduct meriting a punitive damages award. The notion of a crime does not include only intentional or even advertent interferences with important interests;(107)' it also embraces certain forms of grossly(108)' or ordinary negligent conduct,(109)' and, in the case of crimes of strict liability, conduct that does not display even this degree of fault. The result is that there is still a need for some additional concept which delineates the most serious forms of crime.

5.127 ''We would also question a key assumption underlying option 2. This is that there is an extensive range of conduct which merits a punitive damages award, ranging from the highly to the barely culpable; the corollary, it is argued, is that conduct at the lower end of this spectrum should be capable of being insured against, whilst conduct at the higher end of this spectrum should not.

5.128 'The critical point is that even though there is such a range of conduct, a basic minimum threshold of bad conduct must have been reached before an award of punitive damages can properly be made by a court. The aim of an award is the same wherever on the spectrum a particular defendants conduct falls: the conduct is thought to be sufficiently bad to require punishment. If this is so, it is incoherent for the law then to be seen to say: "even though we thought fit to punish you, in reality we are not concerned about whether it will be efficacious, or not, because your conduct was not that bad".

5.129 ''Option 3 would, in its practical impact, be most closely consistent with what may be the existing judicial approach to insurance against criminal punishment.(110)' The policy which might be thought to justify option 3 was well expressed by Denning J in Askey v Golden Wine Co Ltd':

''It is, I think, a principle of our law that the punishment inflicted by a criminal court is personal to the offender, and that the civil courts will not entertain an action by the offender to recover an indemnity against the consequences of that punishment. In every criminal court the punishment is fixed having regard to the personal responsibility of the offender in respect of the offence, to the necessity for deterring him and others from doing the same thing against, to reform him ... All these objections would be nullified if the offender could recover the amount of the fine and costs from another by process of the civil courts.(111)'

5.130 ''We anticipate that any conduct satisfying the test of a deliberate and outrageous disregard of the defendants rights would be conduct which is sufficiently serious to merit a bar within the criminal law, if such conduct were to constitute a criminal offence. Nevertheless, despite this analogy, and the force of the arguments which underlie it, we believe that a more powerful set of counter-arguments (namely, the five reasons set out above for favouring option 1)(112)' entail that a different approach can and must be adopted in relation to punitive damages awarded in civil actions, than is applied to crimes.

5.131 ''Several consultees made some interesting proposals for dealing with insurance against punitive damages in ways which differed from options 1-3. We think it useful and necessary to describe them, and to give some reasons why we ultimately reject them.

''(i)Insurance is permitted only to the extent that there is a shortfall caused by a wrongdoers inability to meet his or her liability

5.132 ''One suggestion(113)' was (in effect) that any insurance cover for punitive damages should be limited to such sums as are necessary to meet a shortfall arising due to the insured wrongdoers inability to pay all or part of any award out of his or her own assets.

5.133 'The merits of this proposal are two-fold. On the one hand, plaintiffs would be certain of having their claims satisfied, in those cases where a defendant is insured. On the other hand, the punitive effect of a punitive damages award would be preserved in an undiminished, or at least less diminished, form. Defendants would, in a greater number of cases, feel an award directly in their own pockets, rather than indirectly through, for example, increased insurance premiums for the future, or the inability to renew previous cover. This might always be so where the defendant (for example, a large profit-making organisation) has sufficient assets to meet a claim, without recourse to an insurance policy.

5.134 'However, this superficially attractive argument raises considerable difficulties. The first problem is that it is not easy to see why a potential insured, if properly advised, would want an insurance policy limited in the way proposed. Under our recommendations, wrongdoers will never be required to pay more than they are able (without undue hardship) to pay.(114)' Thus to apply this insurance against shortfall suggestion would mean that wrongdoers would be no better off if they obtained insurance (because they would still have to meet any punitive award, out of their own pockets, to the extent that they were able to do so). As a result, if properly advised, no-one would want cover for punitive damages, and the net effect would be the same as if the law prohibited cover against punitive damages.(115)'

5.135 'The second problem with this proposal is that it is likely to produce the sort of problematic conflict between insurer and insured wrongdoer which we have already identified.(116)' Insurers would clearly want to argue that the insured-wrongdoer is able to pay the award, thereby reducing the sums which they are obliged to pay under the policy. In contrast, insured-wrongdoers would want to argue that they are unable to pay the award (in full or in part), thereby reducing the sums which they have to pay out of their own pockets. It cannot be desirable to introduce such conflict, with resulting uncertainties and costs, without good reason. As we have already indicated, we doubt whether such a reason exists.

5.136 ''Another suggestion was that insurance should not be permitted, except against vicarious liability.(117)' This might represent the existing common law position, following Lancashire County Council v Municipal Mutual Insurance Ltd':(118)' insurance was held to be permitted against vicarious liability to pay punitive damages, but no final conclusion was reached on the legality of insurance against a personal liability to pay punitive damages.

5.137 'Underlying this proposal is the view that insurance eliminates the punitive and deterrent effect of punitive damages awards, and is therefore (generally) undesirable. Insured wrongdoers do not feel the immediate impact of any award in their own pockets: the primary burden is borne by the insurer. In contrast, permitting insurance against vicarious liability does not of itself serve to frustrate the punitive function of a punitive damages award. The law does not seek to punish the party who is vicariously, rather than primarily, liable; other considerations justify the doctrine of vicarious liability.

5.138 'We recognise the logic of this argument, but for several reasons we are unconvinced of its weight. First, we re-emphasise our doubts that the availability of insurance will wholly or even substantially eliminate the punitive and deterrent effects which may typically be expected of awards of punitive damages.(119)' Secondly, we consider that the reasoning underlying this proposal is inconsistent: vicarious liability also compromises the punitive function of punitive damages awards. Where awards are paid by a primary wrongdoers employer, the wrongdoing-employee escapes direct punishment by the law. He or she does not meet the liability out of his or her own pocket, and the direct punitive effect of an award is replaced by indirect pressures in the form of, for example, contribution or indemnity claims by the employer, or disciplinary action. The position is similar where insurance is permitted against the primary liability: the insured-wrongdoer does not feel the impact of an award directly in his or her own pocket. Accordingly, if the possibility of indirect punishment and/or deterrence is held out as one reason why vicarious liability is acceptable, the same argument ought to have at least some weight when deciding whether insurance should be permitted against a primary liability to pay punitive damages.(120)' Thirdly, insurance and vicarious liability have in common one important rationale. The single most important reason both for permitting insurance and for recognising vicarious liability is the same: to ensure that judgments for punitive damages can be satisfied and therefore that victims of outrageous conduct have a financial reason for claiming punitive damages. To recognise vicarious liability to punitive damages and permit insurance in respect of it, whilst prohibiting insurance against a primary liability to pay punitive damages, ignores this.

5.139 ''A final suggestion was that insurance would be permitted against only a fixed percentage of an award of punitive damages.(121)' The main objections to this proposal are two-fold. The first objection is that it is difficult to select any particular percentage in a non-arbitrary way. This is due to an inherent flexibility in the appropriate balance between ensuring that plaintiffs have a financial reason to claim punitive damages and ensuring effective punishment or deterrence. Whereas the first goal is better served by permitting a higher percentage of an award to be covered by insurance, the second is better served by permitting a lower percentage. The second objection is that, accepting that it is essential for plaintiffs to have a financial reason to claim punitive damages, any percentage chosen would have to be high. This means that there would be very little difference between option 2 and this proposal, so far as the punitive efficacy of punitive damages awards is concerned. This is even more clearly the case given our doubts about how far the full availability of insurance will entail any substantial dilution of the punitive and deterrent effects that may typically be expected of punitive awards.(122)'

''(42) our draft Bill should clarify that insurance against the risk of an award of punitive damages is not against public policy. '(Draft Bill, clause 9(1))

5.141 ''Insurers, of course, remain able to refuse (or in some way limit or impose conditions on) cover for punitive damages awards. In the United States insurers have responded to the availability of exemplary or punitive damages by attempting to exclude them from the scope of their policies.(123)' In this country it may already be difficult to obtain cover for certain types of claim for which exemplary damages are currently available.(124)'

5.142 ''A number of statutes directly or indirectly require liability insurance in certain circumstances.(125)' The areas of activity covered by these schemes are extremely varied. Each clearly requires insurance against a liability to pay compensatory damages in specified circumstances. None expressly requires insurance against a liability to pay exemplary or punitive damages in those circumstances. Could any of these statutes be construed as doing so? Should any of the statutes be so construed?

5.143 'It would be very difficult to argue that any of the existing statutory requirements for insurance were intended, when passed, to extend to a liability to pay exemplary or punitive damages. The primary purpose of compulsory insurance is plaintiff-protection: ensuring that defendants are able (via the insurer) to meet judgments against them. Each existing statutory requirement is such that the dominant aim must have been to ensure that defendants could compensate 'plaintiffs for their injuries.(126)' In some of the statutes, for example, the compulsory insurance provisions are expressed to apply only to a liability which the statute itself establishes, and that statutory liability is a liability to compensate only. In all of the other cases, the compulsory insurance provision applies to an area of activity in which it would be extremely rare (or impossible) on the state of the law, at the time of enactment, for a claim to exemplary damages to succeed.(127)' It would in any case require a rather forced interpretation of the aim of plaintiff-protection, for it plausibly to extend to require that insurance against exemplary or punitive damages be compulsory. For, on the face of it, there is injustice to the plaintiff only if he or she cannot obtain compensation - not if he or she is merely unable to obtain the windfall of an exemplary or punitive damages award.

5.144 'In our view, the fact that the legislature has made insurance compulsory in an area of activity, so as to ensure that claims to compensation can be satisfied, provides no justification for concluding that the relevant statute should in future, in view of the new remedy of punitive damages which we propose, extend to require insurance against liability to that remedy also. Whether this is so must be a matter for the legislature to decide, rather than for resolution on the basis of assumptions about what the enacting legislature might have decided, if the law had then been what we propose it should now be. The decision is pre-eminently a policy decision which is appropriate for the legislature, and not for the courts.

5.145 'We therefore consider that no Act or subordinate legislation should be construed to require insurance against a liability to pay punitive damages. Statutory clarification of this point is essential for both insurers and insured; it cannot be left to ad hoc 'resolution by courts following litigation. We recommend that:

''(43) our draft Bill should ensure that, unless a future enactment expressly or clearly requires insurance against a liability to pay punitive damages, no enactment shall be construed to require it.' (Draft Bill, clause 9(2))

''This recommendation will preclude any argument that any of the existing compulsory insurance statutes extend to punitive damages: none of those statutes expressly cover a liability to pay punitive damages. But it will not prevent a future Act from requiring insurance against such a liability, should the legislature decide that such a requirement is appropriate, provided that it is made clear (by express words) that that is the intention.

5.146 ''At present no claim for exemplary damages survives for the benefit of the estate of a deceased victim of wrongdoing.(128)' This rule can be criticised on a number of grounds,(129)' and repeal was supported by a majority of consultees. We consider that wrongdoers ought to be punished whether or not their victims are alive: a wrongdoer should not escape punishment as a result of a fortuity.(130)' And crucially, the aims of both retribution and deterrence will be furthered by the survival of a punitive damages claim for the benefit of the estate of the victim.

''(44) section 1(2)(a)(i) of the Law Reform (Miscellaneous Provisions) Act 1934 should be repealed and the Act amended so as to allow claims for punitive damages to survive for the benefit of the estate of a deceased victim. '(Draft Bill, clause 14(1)-14(3))

5.148 ''At present an award of exemplary damages can' be claimed from the estate of a deceased wrongdoer.(131)' We think this is the wrong approach.(132)' Unfortunately we have not benefited from the views of any significant number of consultees on this question. Responses dealing with the survival issue almost uniformly dealt with survival for the benefit of the victims estate, and failed specifically to consider the present rules on survival against the wrongdoers estate.(133)' Nevertheless, we have been sufficiently persuaded by the arguments in one clear response on this issue, given by Professor Tettenborn, to feel confident that the existing approach should not stand. Professor Tettenborn writes:

''On the question of survival against the estate of the defendant ... I disagree with the tentative suggestion that exemplary damages should be assimilated to other causes of action. On the point of principle, we are here dealing with punishment; no question of compensation arises. In such a case there seems no need to visit the sins of the parents on the children and the heirs. Suppose a policeman guilty of brutality subsequently dies; it seems inhumane to tell his widow and children that, even though the victims have been fully compensated, they are liable possibly to lose their home in order to satisfy a judgment for exemplary damages. Note too the analogy of criminal law, where I do not think it has ever been seriously suggested that we should introduce the posthumous trial of dead offenders with a view to levying a fine on their estates.

5.149 ''Thus where the wrongdoer who is to be punished is dead, the retributive goal of a punitive award cannot be achieved; only the innocent heirs are punished. It can be argued that there is no unfairness in this, because the estate would have been diminished by the same amount even if the wrongdoer had not died. But in our view this argument is refuted by the plausible scenario described by Professor Tettenborn: in such circumstances there may, on the contrary, be very significant unfairness to the defendants heirs (his family). Furthermore, it is far from clear that a punitive award has any other significant point, and so justification, where the wrongdoer is dead. Individual deterrence offers no argument, for the reason that, having died before the conclusion of an action against him, the wrongdoer cannot act, let alone act wrongfully, in the future. Nor is the argument from general deterrence a strong one. Potential wrongdoers would not usually be any less deterred if the law refused to permit an action to survive against a dead wrongdoer. Such persons will generally expect to be alive, not dead, when an action is brought - and if alive, a claim to punitive damages can be made against them.

''(45) the Law Reform (Miscellaneous Provisions) Act 1934 should be amended in order to prevent punitive damages from being available against a wrongdoers estate.' (Draft Bill, clause 14(1) and 14(3))

5.151 ''Parliament has, as we have already seen,(134)' rarely thought it necessary to authorise exemplary or punitive damages by statutory provision. It has expressly done so in one case,(135)' and arguably done so in another.(165)' We consider that consequential amendments are required to each of these statutes.

''(a)Reserve and Auxiliary Forces (Protection of Civil Interests) Act 1951, s 13(2)

5.152 ''The first amendment needed is to section 13(2) of the Reserve and Auxiliary Forces (Protection of Civil Interests) Act 1951, which authorises an award of "exemplary damages". It would be undesirable if that power could be construed as authorising the courts to award a punitive sum of damages which was governed by principles other than those stated in our draft Bill.(137)' We therefore recommend that:

''(46) section 13(2) of the Reserve and Auxiliary Forces (Protection of Civil Interests) Act 1951 should be amended, so that, in place of exemplary damages, it authorises an award of punitive damages to which our Act applies. '(Draft Bill, clause 14(4))

5.153 ''The second set of amendments is to sections 97(2), 191J and 229(3) of the Copyright, Designs and Patents 1988. These sections provide, respectively, for an award of additional damages for infringement of copyright, performers property rights and design right. We have seen that the proper characterisation of additional damages is controversial.(138)' In our view the appropriate course is to repeal sections 97(2), 191J and 229(3), and we so recommend:

''(47) sections 97(2), 191J and 229(3) of the Copyright, Designs & Patents Act 1988 should be repealed.' (Draft Bill, clause 14(5))

5.154 ''We consider it necessary to take this step for several reasons. Repeal of those sections will eliminate the uncertainty which has surrounded additional damages; the remedy (whatever its proper characterisation) shall thereafter be unavailable. But this will not leave any significant lacunae in the laws protection of intellectual property rights.

5.155 'We have recommended(139)' that punitive damages should be available for a statutory civil wrong if an award would be consistent with the policy of the statute in question. All of the wrongs which are affected by recommendation (47) fall into this category. And, we firmly believe, it would be consistent with the policy of the Copyright, Designs and Patents Act 1988 if punitive damages could be awarded in respect of those wrongs.(140)' To the extent that additional damages are punitive damages, therefore, there will be no legitimate additional role for them, if our recommendations are implemented. For to allow punitive additional damages to continue to exist would involve accepting, as we do not, that a punitive award can be made which is not subject to the limitations imposed by our Act.

5.156 'Even if additional damages are best viewed as compensatory in nature(141)' (which we doubt) we can nevertheless see no convincing reason for retaining them. In our Consultation Paper we observed that:

''In so far as s 96(2) of the Act provides a general remedy for copyright infringement of damages which are at large as well as a remedy of account ... it is difficult to see the role of s 97(2) if exemplary damages are not permitted by it.(142)'

''There is, we believe, no reason why aggravated damages (that is, damages for mental distress) should not be capable of being awarded, apart from sections 97(2), 191J and 229(3), for infringement of copyright, performers property rights or design right. This seems to have been assumed in at least two cases.(143)' Nor is there any necessary bar to judicial development of exceptions to the usual rules of remoteness as they exist at common law.(144)' In Smith New Court Securities Ltd v Scrimgeour Vickers (Asset Management) Ltd'(145)' 'Lord Steyn 'justified the special deceit rules in terms which prima facie 'also justify wider rules in relation to intentional wrongdoing generally.(146)' It is surely not beyond the capacity of the common law to hold, if necessary, that flagrant infringements of copyright merit an increased measure of compensatory damages.

5.157 ''Nor will the abolition of additional damages lead to any lacunae in restitutionary remedies for these intellectual property wrongs. It is true that a court is specifically directed to take into account "any benefit shown to have accrued to the defendant by reason of the infringement" in deciding whether or not to award additional damages. But this is certainly not a decisive indication that additional damages have a restitutionary rather than a punitive aim.(147)' And we are unaware of any judicial authority, or of any support in the legislative history of section 97(2), for the view that additional damages are restitutionary in aim.(148)' Even if additional damages could include restitutionary damages, abolition would not leave lacunae because an account of profits will remain available to victims of such wrongs.(149)' Moreover, if the defendants conduct has shown a deliberate and outrageous disregard of the plaintiffs rights, the victim may be entitled to claim restitutionary damages under our Act.(150)'

5.158 'We also think it important that several intellectual property lawyers have emphasised to us how anomalous it is that the special remedy of additional damages is only available for a limited number of intellectual property torts. Our proposals have the merit of making punitive damages available for all such wrongs, (although if an intellectual property tort arises under an Act, this is only so if such an award would be consistent with the policy of the Act in question).(151)'

5.159 ''Insurers have expressed to us the concern that increases in the quantum of punitive damages should not apply in respect of insurance cover which they have already given, and therefore the hope that our Bill will not apply retrospectively. In order to accommodate this concern, we recommend that:

''(48) our draft Bill should provide that nothing in it applies to causes of action which accrue before its commencement. '(Draft Bill, clause 16(1))

5.160 ''This means that where a cause of action accrued before 'commencement, the old law of exemplary damages, as defined by (in particular) the categories test and the cause of action test will continue to apply to a claim for damages in respect of it. Where, however, a cause of action accrues after commencement, the expanded remedy of punitive damages, as defined by our Act, will apply.

(1) Aggravated, Exemplary and Restitutionary Damages (1993) Consultation Paper No 132, paras 6.9 and 8.8. See also paras 6.14 and 8.9.

(3) See on torts, Clerk & Lindsell on Torts(17th ed, 1995) ch 31; generally, see A McGee, Limitation Periods (2nd ed, 1994).

(5) See eg Clerk & Lindsell on Torts(17th ed, 1995) paras 3.33-3.56; on contractual limitation or exclusion clauses generally, see Chitty on Contracts (27th ed, 1994; first cumulative supplement, 1996).

(6) See Dicey & Morris on the Conflict of Laws (12th ed, 1993; fourth cumulative supplement, 1997) vols 1 and 2.

(8) See eg Clerk & Lindsell on Torts(17th ed, 1995) para 27.25; McGregor on Damages (15th ed, 1988) ch 14.

(9) See eg Clerk & Lindsell on Torts(17th ed, 1995) para 27.16; McGregor on Damages (15th ed, 1988) ch 13.

(11) See eg I F Fletcher, The Law of Insolvency (2nd ed 1996) ch 9 (personal insolvency) and ch 28 (corporate insolvency).

(12) See eg I F Fletcher, The Law of Insolvency (2nd ed, 1996) pp 288-299 (personal insolvency) and pp 606-613 (corporate insolvency). We have given some thought to the question whether unliquidated claims to, or judgment debts for, punitive damages should be capable of being proved on personal or corporate insolvency, and if so, how such claims should rank. At present they rank as ordinary unsecured claims, and can be proved in the same way as any other claim to damages, whether liquidated or unliquidated. At first sight, it might be thought unfair to other (innocent) unsecured creditors if they were to receive less because the defendant's assets also had to be used to satisfy what many regard as 'windfall' claims to punitive damages. But on reflection, we do not think it appropriate to consider this issue further in this Report. It was not raised in Consultation Paper No 132 and no consultees alerted us to any problems with the current law in this area. (We would also observe that, in any case, a claim to punitive damages merely ranks as an ordinary unsecured claim; it ranks alongside other such claims, and does not take priority over them). Although we say nothing further on this issue, we do draw it to the attention of the Lord Chancellor, who has statutory responsibility for making rules as to inter alia what debts are provable on personal or corporate insolvency and how those debts rank, and to the specialist bodies (in particular the Insolvency Service and the Insolvency Rules Committee) which assist him in this task. Cf US law: L Schleuter and K Redden,Punitive Damages (3rd ed, 1995) § 19.5 indicate that under the Bankruptcy Code, exemplary or punitive damages can be proven on bankruptcy, but are subordinated to the payment of all other types of claim.

(16) In Aggravated, Exemplary and Restitutionary Damages (1993) Consultation Paper No 132, para 6.47, we provisionally supported the view that, as at present, no detailed inquiry into the defendants finances should be undertaken. There was a mixed response to this from consultees.

(17) We do not rule out the possibility that some plaintiffs may wish to bring an action simply to have their rights vindicated.

(18) This option was briefly discussed, without a provisional view being reached, in Aggravated, Exemplary and Restitutionary Damages (1993) Consultation Paper No 132, para 6.38. We have chosen 33% rather than a higher percentage - which might make diversion more cost-effective - because we are concerned about the potential implications for the pleading of punitive damages of 50% diversion. We believe that only the plaintiff should be able to plead punitive damages - and not, for example, in addition some person or body like the Attorney-General acting on behalf of the general public. To choose a 50-50 division might have the unfortunate effect of suggesting that the victim and the public have

(19) Arbitrariness is particularly inevitable, given that what is cost-effective cannot be determined without knowing what the likely number and scale of awards will be; such information will not be ascertainable with accuracy until any scheme has been in place for some time; and the degree to which any figure fixed upon is arbitrary will depend on the very variable behaviour of litigants.

(21) A possible approach (which we do not prefer) would be to leave the enforcement of punitive damages to the court of its own motion, by, for example, appointing a court officer to supervise compliance.

(23) Such diversion would be most unlikely to be cost-effective, given that awards against the state may be anticipated to be the greatest single category.

(25) Legislative intervention, proposed or enacted, has included: caps; a first comer gets all rule; a rule which credits a defendant with prior punitive payments; a rule which permits punitive damages class actions at the instance of a defendant; the consolidation of all multiple punitive claims; the use of an injunction against the enforcement of individual punitive judgments until they could all be consolidated for a single administration in a single court; and the bifurcation or trifurcation of trials, to separate liability and damages issues from punitive proof. The courts have generally recognised multiple punitive liability.

(27) There will obviously be no bar to claims to punitive damages which are founded on conduct other than that which was the basis for the claim in the first successful action.

(28) The question of whether the conduct alleged by P1 as the basis for a claim to punitive damages from D is the same as conduct alleged by P2 (or P3 ...), is, we believe, one best left to courts to resolve.

(29) For example, each multiple plaintiff must show that the defendant committed a tort, equitable wrong (as defined) or statutory wrong (as defined) against him or her, and that the defendants conduct in so doing, or subsequent to the wrong, showed a deliberate and outrageous disregard for that plaintiffs rights.

(31) Indeed, it also includes a liability to multiple plaintiffs who were awarded, eg, compensation, in an action

(33) See, in particular, RSC O 15, r 4(1) (joinder of parties) and RSC O 4, r 9 (consolidation); see for guidance on the use of these powers in group actions, the Supreme Court Procedure Committees

(41) Cf the effects of contribution. If contribution is possible, because there are one or more other concurrent tortfeasors who are solvent and similarly culpable, this will only further reduce the likely liability of each of those tortfeasors.

(48) Aggravated, Exemplary and Restitutionary Damages (1993) Consultation Paper No 132, paras 6.45 and 8.17(f), proposing the Irish Civil Liability Act 1961 as a model for reform. Our clause (8(1) of the draft Bill) aims to achieve the same result.

(49) The availability of contribution in respect of a liability to compensate will be unaffected. See paras 5.206-5.207 and recommendation (36) below.

(57) The proposition that a liability to punitive damages should be several means that punitive damages should only be awarded where they are available because of, and are assessed by reference to, the defendants conduct. The corollary is that any award so available and assessed should be payable only, and in full, by

(58) This may be sufficiently clear from the general proposition in our draft Bill (clause 8(1)) that a liability to pay punitive damages is several only. Nevertheless, subject to this, the wording of s 1(1) and s 6(1) of 1978 Act is broad enough to entitle a person to claim contribution in respect of a liability to pay punitive""

(60) The question was discussed in Aggravated, Exemplary and Restitutionary Damages (1993) Consultation Paper No 132, paras 6.42-6.44, without any provisional view being reached.

(62) Whether directly, by initial proceedings against the employer, or less directly, by way of contribution or indemnity claims between employer and employee.

(64) For a thorough consideration and criticism of the many different arguments which have been advanced in support of vicarious liability, see P S Atiyah,

(68) For economic theory supporting this conclusion, see A Ogus, "Exemplary Damages and Economic Analysis" in K Hawkins (ed),

(69) An employee dismissed for such a reason would be most unlikely, for example, to be able to obtain favourable references from his or her former employer.

(70) In this respect it is interesting to observe that the statement of claim in ""

(75) One concept would apply to the claim to compensation for the wrong; another concept would apply to the claim for punitive damages.

(78) Sex Discrimination Act 1975, s 41; Race Relations Act 1976, s 32; Disability Discrimination Act 1995, s 58. The phrase "in the course of employment" is to be construed less technically and restrictively (see

(79) In particular, the Crown Proceedings Act 1947, s 2(1)(a) (Crown); Police Act 1996, s 88(1) (chief officer of police); Police Act 1997, s 42(1) (Director General of the National Criminal Intelligence Service), s86(1) (Director General of the National Crime Squad).

(80) The statutory provisions which extend the doctrine of vicarious liability to (eg) the Crown or to chief officers of police (referred to above), only deal with vicarious liability for

(81) There is little authority for vicarious liability for equitable wrongs: see para 4.102, n 228 above.

(82) In particular: (i) did the employee commit a wrong for which punitive damages may be awarded?; (ii) did the employees conduct demonstrate a deliberate and outrageous disregard of the plaintiffs rights?; (iii) are other remedies or sanctions inadequate to punish the employee for his conduct?; (iv) what sum of punitive damages should be awarded in order to punish the employee for his or her conduct, taking account of the various principles and factors which our Bill requires a court to take into account?

(85) This issue was discussed, without any provisional view being reached, in Aggravated, Exemplary and Restitutionary Damages (1993) Consultation Paper No 132, paras 3.111-3.112 and 6.37.

(90) See the discussion in Aggravated, Exemplary and Restitutionary Damages (1993) Consultation Paper No 132, para 6.39-6.41; see also, on the approach of United States jurisdictions,

(91) Aggravated, Exemplary and Restitutionary Damages (1993) Consultation Paper No 132, paras 6.39-6.41.

(104) We recognise that permitting insurance will not remove this problem: it could arise in any case where defendants did not insure or could not insure, either because the premiums demanded were too high, or because the insurers excluded punitive awards from the scope of their policies. Nevertheless, if insurance is permitted, the conflict is at the very least not inevitable.

(107) But many more serious crimes do require such a higher degree of fault or culpability.

(108) See eg the Road Traffic Act 1988, s 1 and s 2 (causing death by dangerous driving; dangerous driving - for the meaning of dangerous, see s 2A).

(109) See eg the Road Traffic Act 1988, s 3 and s 3A (careless or inconsiderate driving; causing death by careless or inconsiderate driving).

(110) See para 4.108 above. The approach to contracts of indemnity is also applicable to other forms of indemnity (eg by way of a tort action for damages) in respect of fines paid by way of punishment, and even against the adverse financial implications of conviction (eg loss of business profits).

(115) A similar objection can be raised to the suggestion of (eg the Association of Personal Injury Lawyers) that insurers should be required to meet any liability to pay punitive damages in full, but should be given a right of recourse against the insured.

(117) For example: P Cane, 1 Pump Court (R Latham), and Sinclair Roche & Temperley.

(120) It is possible to argue that by allowing insurance against vicarious liability, one only further weakens the likely indirect pressure on wrongdoing-employees: if the immediate burden of awards which their employers must pay is borne by their insurers, they have less of an incentive to discipline their employees.

(123) See Aggravated, Exemplary and Restitutionary Damages (1993) Consultation Paper No 132, para 6.41, referring to research conducted for the Law Commission; for a general overview of the approach of the courts to such attempts, see L Schlueter and K Redden,

(124) On consultation, Peter Carter-Ruck observed that most insurance policies for libel exclude liability for publications found to be malicious, and the Association of Chief Police Officers indicated that police officers are finding it difficult to obtain insurance cover for exemplary damages, notwithstanding that in almost every case in which police are involved, there is a risk of an exemplary damages award.

(125) Road Traffic Act 1988, s 143; Nuclear Installations Act 1965, s 19; Employers Liability (Compulsory Insurance) Act 1969; Merchant Shipping Act 1995, s 163; Riding Establishments Act 1964, s 1(4A)(d); Civil Aviation (Licensing) Regulations 1964; Insurance Brokers (Registration) Act 1977, s 12, and the rules made pursuant thereto; Dangerous Wild Animals Act 1976, s 1(6)(a)(iv); Solicitors Act 1974, s 37, and the rules made pursuant thereto; Credit Unions Act 1979, s 15; Estate Agents Act 1979, s 16.

(126) Thus, in many cases, we find parallel compensation funds established (eg Merchant Shipping Act 1995, Ch IV; Nuclear Installations Act 1965, s 18; rules made pursuant to the Insurance Brokers (Registration) Act 1977 (see SI 1987 No 1496 and SI 1990 No 2461); Solicitors Act 1974, s 36); expressly-created direct rights of action for victims against insurers (Merchant Shipping Act 1995, s 165; see also Road Traffic Act, s 153); and terms in the insurance contract being rendered ineffective as against victims (Road Traffic Act 1988, s 148).

(127) For example, exemplary damages have never been awarded for the tort of negligence, yet this would be the main basis for a claim against motorists (covered by the Road Traffic Act 1988) or against employers by their employees (covered by the Employers Liability (Compulsory Insurance) Act 1969). The same reasoning applies to professional indemnity insurance. Exemplary damages cannot be awarded for negligence, breach of contract, deceit or pre-contractual misrepresentations actionable under s 2(1) of the Misrepresentation Act 1967. In addition to the unlikelihood of there being a cause of action for which exemplary damages could be claimed, there is the difficulty of fitting claims within one of the three

(128) Law Reform (Miscellaneous Provisions) Act 1934, s 1(2)(a)(i). This is also the prevailing approach in other major Commonwealth jurisdictions: see para 4.106 above.

(129) See Consultation Paper No 132, paras 3.108-3.110, 6.36; and see, for similar criticisms, the Ontario Law Reform Commission,

(130) The Association of Personal Injury Lawyers pertinently pointed out that only a politician or lawyer would tolerate the suggestion that if a person maims or cripples someone then punitive damages may be awarded against him but, if he goes further and kills the victim, then he is free and no question of punitive damages arises.

(131) This is also the prevailing approach in other major Commonwealth jurisdictions: see para 4.107 above.

(132) This view is supported by the conclusions of the Ontario Law Reform Commission, in its

(133)Some ambiguity in the question which we put to consultees is very probably to blame for this.

(135) Reserve & Auxiliary Forces (Protection of Civil Interests) Act 1951, s 13(2).

(137) For example, the power might not be subject to the requirement that the defendant has shown a deliberate and outrageous disregard of the plaintiffs rights, or to the if, but only if test, or to the various other principles which govern the availability and assessment of punitive damages under our Act.

(140) Apart from the fact that Parliament apparently did consider that a punitive remedy was necessary for these wrongs (and so expressly provided for the remedy of additional damages), there is also the fact that the Act provides (by ss 96(2), 191J and 229(2)) that in an action for any of these wrongs:

(142) Aggravated, Exemplary and Restitutionary Damages (1993) Consultation Paper No 132, para 3.54.

(147) See, in particular, the discussion of category 2 exemplary damages at paras 4.16-4.19 above.

(148) See the discussion at paras 4.21-4.22, which indicates that judicial disagreement about the characterisation of additional damages has been a disagreement about whether they are best viewed as authorising awards of exemplary damages, or a higher measure of compensation than could be obtained on ordinary principles.

(150) See clause 12 of the draft Bill. Restitutionary damages will be available for wrongs which arise under an Act for which a person may recover compensation or damages where (i) the defendants conduct showed a deliberate and outrageous disregard of the plaintiffs rights; and (ii) an award of restitutionary damages would be consistent with the policy of that Act.

The Law Commission