2.    Unlike the position in some other common law countries, there are no specific rules of court in England and Wales for multi-party actions. This causes difficulties when actions involving many parties are brought. In addition to the existing procedures being difficult to use, they have proved disproportionately costly. It is now generally recognised, by judges, practitioners and consumer representatives, that there is a need for a new approach both in relation to court procedures and legal aid. The new procedures should achieve the following objectives:

(a) provide access to justice where large numbers of people have been affected by another's conduct, but individual loss is so small that it makes an individual action economically unviable;

(b) provide expeditious, effective and proportionate methods of resolving cases, where individual damages are large enough to justify individual action but where the number of claimants and the nature of the issues involved mean that the cases cannot be managed satisfactorily in accordance with normal procedure;

(c) achieve a balance between the normal rights of claimants and defendants, to pursue and defend cases individually, and the interests of a group of parties to litigate the action as a whole in an effective manner.

3.    In 1992 the Court of Appeal said that there might be a strong case for legislation to provide a jurisdictional structure for the collation and resolution of mass product liability claims (Nash v Eli Lilly & Co. [1993] 4 All ER, at p.409). The Legal Aid Board has called for new procedures tailored to multi-party litigation, which will emphasise the central issues rather than investigating every individual claim in detail. Other common law jurisdictions have similarly found the need to enact statutory provision for multi-party actions. Nearer to home, the report of the Scottish Law Commission has been completed and is being published.

4.    During the second stage of the Inquiry, a working party of the Civil Litigation Committee of the Law Society has prepared a report. The working party included practitioners who act regularly for claimants and for defendants. On nearly all issues the working party reached a consensus and a number of detailed recommendations for change were formulated. These included a new rule which deals comprehensively with the conduct of group actions from their initiation to judgment or settlement. Their recommendations are especially valuable because they have applied the philosophy of case management, espoused in the interim report, to the problem (Group Actions Made Easier, A Report of the Law Society's Civil Litigation Committee, September 1995).

5.    The Inquiry published an issues paper raising a number of further questions in January 1996, and held a seminar involving many of those practising in this area of litigation and representatives of claimants and defendants. I have received 55 responses to the issues paper from practitioners, judges and others who were generous in offering lessons from their own experience. I have also benefited from the deliberations of a small working party, and I am indebted to the Inquiry's academic consultant, Professor Ross Cranston, for a detailed commentary on the experience in other jurisdictions. This, most notably in the United States, draws attention to problems which should be taken into account in developing new multi-party rules in England and Wales.

6.    In this area of litigation more than any other my examination of the problems does not pretend to present the final answer but merely to try to be the next step forward in a lively debate within which parties and judges are hammering out better ways of managing the unmanageable.

Existing rules
7.    Although the existing rules of court provide means of dealing with multi-party actions, they were not drafted with group actions in mind and therefore none has provided a sufficient answer to the problems they create. Representative actions are provided for by RSC Order 15, rule 12 but the experience here and in comparable jurisdictions is that there are definite limits to the weight the rule can bear. Cases can also be joined or consolidated under Order 15, rule 4 and Order 4, rule 9(1). But consolidation deals with situations where actions have already been begun, and it is better that multi-party litigation be dealt with on a collective basis before then, and joinder is not satisfactory where the interests of claimants differ.

The problems with multi-party litigation
8.    The problems of cost and delay identified in my interim report are magnified in the context of group actions. Cases take on a life of their own and there is insufficient independent continuing consideration of whether the cost of the litigation is justified by what is at stake. There is a great risk that actions involving large numbers of claimants will become management or organisation driven because of the sheer scale of the numbers involved. Decisions which in a single case might have a small negative impact, when multiplied many hundreds or even thousands of times, can produce waste of effort and resources on a large scale. In addition the complexity and intractability of the intrinsic subject matter can generate major discovery exercises and escalating use of experts to an even greater extent than in ordinary litigation. The large numbers of potential claimants can mean that substantial cost becomes the norm.

9.    The positions of claimants and defendants appear inevitably to become polarised over strategy: the claimants' wish to broadly focus on the common or generic issues, the defendants' wish to identify and investigate each individual case. A confrontational climate develops which fuels this divergence of views.

10.    The differing interests of group members, even where there are substantial common issues or interests, give rise to difficulties in establishing generic issues applicable to the entire group, maintaining overall progress of the case and achieving settlement for the whole group. Separate claimants with separate representatives may find themselves at odds with each other or unnecessarily duplicating effort and expense. Although most attention is generally given to the problems of high profile 'disaster' cases, similar problems arise in relation to a wide variety of cases. In housing cases, for example, tenants in high rise blocks may have different objectives from those in low rise housing; long lessees from weekly tenants; tenants who wish repairs to be executed from those who seek rehousing.

11.    The desire of defendants to know the scale of the action they face leads to the setting of cut-off dates which in turn can cause the swamping of valid claims with weak or hopeless claims. Defendants may suffer from the adverse publicity resulting from the number of potential claimants and may have to bear the expenses of the individual investigation of such cases. There are also problems arising from the need to protect the rights of individuals who are not part of the group and to balance this need against the desire of the defendants for finality. Settlement of some cases or settlement without any court supervision of lead or test cases may undermine the viability of the group as a whole.

12.    In larger actions, the costs may be so enormous and so uncertain that only those eligible for legal aid can contemplate involvement. The escalation of costs in such cases can put the initial cost benefit analysis at risk. In more modest actions, the cost is more proportionate but current legal aid and court arrangements do not always contribute to the most cost effective resolution of the issues. The present rules on legal aid funding appear to rule out the funding of representative actions and yet these may provide a more cost effective solution than litigating a group of many individual cases. Lack of clarity about cost sharing arrangements and what costs are recoverable create significant problems on taxation.

13.    In the absence of legislative support, courts have had to tackle these problems pragmatically, making decisions on a creative and improvised basis with regard to cut-off dates, on how investigation should be conducted, on whether to process some or all cases in the group and on costs. While many judges have risen to the challenge, they themselves have indicated the need for a clearer framework in which to operate.

A new framework
14.    The Law Society's Working Party has recommended that multi-party actions should be managed from the outset. Multi-party actions, of whatever description, will almost invariably merit the full hands-on judicial control which I am recommending for the most complex cases. The system of case management which I propose generally will provide judicial scrutiny at the stage when a defence is filed and appropriate handling thereafter. But in multi-party actions there is a need for the court to exercise control at a much earlier stage. Special arrangements will be required.

15.    My proposals are designed to ensure that the court is notified of the existence of potential multi-party actions as early as possible. At this early stage no immediate decision is required but from then on the judge will be required to take a succession of decisions, all of which will impact on the successful handling of the actions. In this chapter I set out the factors which should be taken into account in relation to three main aspects:

(a) the initial stages, involving the application for and certification of a multi-party situation (MPS), appointment of a designated judge and the arrangements for lead representation;

(b) the strategic priorities for court management, including the definition of the group, the establishment of a register, the need to provide an effective filter, costs; and

(c) protecting the interests of litigants through the court's oversight of lawyers, the appointment of a trustee, the court's approval of settlements.

The initial stages
16.    The earlier the court exercises control in a potential multi-party action the better chance of managing the case to a satisfactory resolution. Other jurisdictions have achieved this by requiring certification of a group or class action where there is an identifiable class or a specified number of persons, and the claims give rise to common issues of fact and law and where handling them together appears to the court to provide the best and the most practicable approach. The disadvantage of the solution usually adopted in other jurisdictions is that there may be many claimants with similar complaints but their claims may be more satisfactorily dealt with, at least in part, in separate proceedings. In this situation, it is likely that a group action will not be certified even though the case would benefit from collective management by the court.

17.    Another approach is to stipulate that for the application of multi-party treatment the claims should give rise to common but not necessarily identical issues. The Law Society's rule extends this concept of flexibility. It recognises that there are clear advantages in drawing together claims which may be in some way related. I would wish to go further and to make it clear that cases that have been drawn together could be dealt with in different ways.

A multi-party situation
18.    To achieve this I recommend that where proceedings will or may require collective treatment to a greater or lesser degree, provision should be made for a multi-party situation (MPS) to be established. This should be achieved with minimum expense by providing that the parties or the Legal Aid Board should make an application to the court which contains a formal declaration that the action meets the criteria for a multi-party situation and the grounds to demonstrate this. There should also be a power for the court itself to initiate or encourage an application. The application will be an executive act. The simplicity and lack of expense will encourage the proposed parties not to delay such a request, thus ensuring that the matter is brought to the attention of the court at the earliest opportunity.

19.    The criteria to be met suggested by the Law Society's working party (draft rule 1.1) are:

20.    I would broadly follow this suggestion, subject to two points. First, the minimum number of ten parties should not be written into the rules but be regarded simply as a guide. Especially in local cases, five may be a sufficient number. Secondly, the common issues need not necessarily predominate over issues affecting only individuals. All that is required is that the court is satisfied that the group will be sufficiently numerous and homogenous for the cases within the MPS to be more viable if there is a collective approach than if they are handled individually.

21.    The MPS will provide maximum flexibility in that it may be proposed that parts of the proceedings are common to some or all of the claimants, and other parts are limited to individual claimants. In addition, individual and common proceedings should take place in parallel or the individual proceedings should take place in advance of or following the common proceedings. One MPS could accommodate the common tools used for disposal of multi-party proceedings, namely test or lead cases, and preliminary or common issues. It could also accommodate a representative action. All actions relating to the MPS could be stayed with their claimants fully protected as to limitation, at minimum cost and without the action being swamped by an influx of new claimants. An MPS will be a suitable framework for handling all the different types of multi-party actions common in this jurisdiction from local housing and environmental actions, consumer cases, financial actions such as the Lloyds litigation, single 'one-off' disasters and large scale complex environmental actions and product liability actions, including pharmaceutical and medical cases. The possible options for dealing with cases within the MPS are explored later in this chapter.

22.    The subsequent procedure for the initial stages of the MPS following the application will be similar to that outlined in the Law Society's rule: a judge will need to decide whether an MPS which has been established should be certified. If it is, then a managing judge should be appointed and should have control of all proceedings arising out of the cases within the MPS. He will need to make decisions about notification of the action, lead lawyers, arrangements for representing the interests of the group, and on how costs will be treated. I deal with these detailed arrangements below.

Joining the MPS
23.    Individual claimants would be able to participate in the application for the MPS by entering their names on a register, as suggested by the Law Society (paragraph 6.10.1-6.10.6) rather than by issue of a separate application for each possible action. Joining the register in this way would, after notification to the proposed defendants of the application for the MPS, suspend the operation of the Limitation Act. While this helps those who are on the register, it does not provide equivalent protection for those in the broader class who have not yet joined the register and the managing judge will therefore need to consider at an early stage the best means of achieving this. I consider this also in paragraph 45 below.

Certification
24.    Certification is confirmation by the court that the criteria for the MPS have been met, so that cases can be treated within it and appropriate orders made and procedures applied. The court will arrange to consider certification after the initial application. If necessary it will ask for further information to enable it to reach a decision. The period for fixing a hearing may vary depending on the scale and complexity of the subject matter of the cases within the MPS but should be no more than three months. Parties may request an earlier hearing.

25.    There is no need for the court to take a view of the merits at the certification stage. The discussion paper on multi-party actions of the Scottish Law Commission (Multi-Party Actions, Discussion Paper No.98, p.184) describes the problems that consideration of the merits would involve for both the court and the parties:

26.    If the certificate were refused then a date for the determination of the MPS would be fixed. If a party wanted to continue proceedings the party would have to file a claim prior to that date to avoid the action coming to an end. A power to decertify is also needed because the situation can change so it is subsequently found that a multi-party situation is no longer appropriate.

Appointment of a managing judge
27.    I have proposed generally that complex cases requiring full hands-on judicial control should be assigned to a single judge. This is to ensure continuity of decision making and will be of particular importance in cases involving complex technical subject matter. The Law Society's Working Party similarly recommended the appointment of a designated High Court judge with power to transfer the proceedings to a designated Circuit judge if "damages were likely to be modest and/or the litigation has a particular connection with a given locality." The appointment of an alternate judge was also recommended. All of this should be done either at or immediately after the certification stage.

28.    Those responding to the issues paper have emphasised that it would be helpful to have arrangements for handling multi-party actions not only in the High Court in London but at courts elsewhere in the country although there is concern that it may not be possible to provide a single managing judge outside London. However, experience over the last year has demonstrated that it is possible to achieve this, particularly at major trial centres on Circuit or by coming to other arrangements such as the transfer of the multi-party estate litigation in Hackney, London to an Official Referee for management throughout the life of the action. I consider that it is important for lower value or local cases to be tried locally at appropriate courts and by either a High Court or Circuit judge. In making these arrangements it is important that the managing judge is appointed as soon as possible following certification and that the judge will be available throughout the life of the action.

29.    The Law Society's working party recommended that the managing judge should appoint a designated Master or district judge. Experience of previous multi-party actions indicates that the judicial workload may be heavy and involve considerable time which it may be difficult for full-time Masters and district judges to provide alongside their other responsibilities. In these circumstances I see benefit in the temporary appointment of a deputy Master or deputy district judge, drawn from those practitioners who already have considerable experience of such litigation. There may also be a role in heavy, complex cases for a law clerk, as I recommend in relation to other complex cases in chapter 8 on the supporting structure.

30.    At this stage, so early in the proceedings, it will be very difficult for any appointed judge to have reached the same stage of familiarity with the subject matter as the claimants and defendants. This problem was specifically discussed at the multi-party action seminar in February 1996 and it was generally agreed that it would be helpful for the judge to have background material made available before he makes any key decisions so that those decisions are based on a reasonable familiarity with the background. Ideally the information made available to the judge should be in the form of a joint submission prepared by the parties and including the background facts, stated jointly, in so far as they are agreed, or the different versions so far as they are not agreed, and a reading list, agreed as far as possible, together with instructions on technical matters.

Arrangements for lead representation
31.    The Law Society's working party suggested that the court should have power to approve the lead lawyer for the group. In privately funded actions, private litigants will normally organise themselves efficiently. In legal aid cases the Legal Aid Board has special arrangements for this purpose. The court's responsibility is not to ensure that the legal services are adequate but to ensure the efficient conduct of the litigation. As this might be hindered by disagreements about lead lawyers in a mixed legal aid and private case such as tenants on a housing estate, or in a private case alone, where parties did not agree, the court should have a residual power to approve the lead lawyer if a difficulty arises. I deal later (paragraphs 70 - 74) with the court's general oversight of lawyers.

Court management: strategic priorities
32.    The Law Society's working party recommended that the designated judge in a multi-party action should be given wide ranging powers to control the litigation and to ensure that it is expeditiously and economically progressed. The need for imagination and creativity in dealing with such litigation is attested to by every judge who has tried such a case. The kernel of the problem is the claimant's desire to pursue generic or common issues and the defendant's equally strong wish to investigate every individual case. There are three basic matters which a judge is almost invariably going to have to start to tackle at the certification stage and immediately after in an MPS.

(a) Deciding whether there are generic issues present and whether they can be effectively decided within the MPS (this bears particularly on the composition and identity of the proposed beneficiaries of the proceedings).

(b) Deciding whether there are issues applicable to certain individuals which need to be determined separately as to those individual claims; and if so to establish machinery for that purpose.

(c) Deciding the order in which the issues identified at (a) and (b) are to be determined.

33.    It is likely to be the case that the judge will usually need to treat (a) as the priority. On the whole, this is likely to be the most rational and economic way of working through the case. But this cannot be a rigid rule. There may be cases where, for example, it will be possible to establish, at relatively limited cost, that there are so few cases in the group in which damage can be proved that it is simply not worthwhile going into the generic issues in any detail.

34.    In many cases, testing the likely viability of a sufficient number of individual cases cannot fairly be postponed until resolution of the generic issues is completed. This is because of the interdependence of the generic and individual issues. The latter shape the former. The cost-benefit justification of the proceedings depends on an adequate number of sufficiently promising cases. And, bearing in mind the adverse effects of a group action on defendants, it is necessary as a matter of basic justice to which they too are entitled.

35.    This does not of course mean that examination of each case is required initially. More selective methods can be used. Although defendants traditionally oppose any selection of lead cases, there is a growing recognition that statistically valid samples of the wider group may be helpful in establishing criteria which individuals must meet to join the action. But above all, consideration of individual cases must not be allowed to paralyse overall progress of the group action.

36.    At this early point the managing judge needs to be pro-active in addressing various key matters with the parties. Some of these will be decisions common to all complex litigation: identifying main and preliminary issues; drawing up a strategy for disclosure, for further investigative work and for the use of expert evidence; establishing a timetable. Others are specific to multi-party actions:

(a) definition of the group;

(b) considering the utility of sub-groups, lead cases or sampling;

(c) considering whether the MPS should be managed on an 'opt-out' basis;

(d) arrangements for giving notice of the action;

(e) establishing a filter by agreeing with the parties the diagnostic or other criteria to facilitate the identification of valid claims and the early elimination of weak or hopeless claims;

(f) determining the approach to costs.

37.    There may be value in the court adopting a less formal approach to proceedings at this stage in order to encourage a more co-operative atmosphere of mutual endeavour to find the best ways of resolving the problems ahead. I understand that at the outset of the Lloyds litigation the court held an informal meeting with interested parties to identify the categories of case involved and to receive information so as to enable the court to apply management techniques.

Definition of the group
38.    In some actions, the claimant group will be already well defined. In a transport disaster there will be a finite group. In a housing case, the group will be the tenants of the estate or lessees of a block of flats and although sub-groups may be helpful there will be a finite number of claimants. In other cases there will be a potential group defined by its circumstances; for instance all those within a specific geographical area in an environment case, or in medical cases, all those treated over a specific period. In some cases the potential group may be very numerous. In each case the judge will need to decide on the most efficient way of bringing potential claimants into the action, on the stage at which this should be done and whether it is appropriate to do this before or after examination of issues of principle or some of the generic issues common to all potential claimants. Clearly, it is pointless establishing a register for a large number of potential claimants if a decision on a key issue of liability or causation might determine the action at an early stage.

Sub-groups, lead cases and sampling
39.    If there is already a substantial number of claims the court can proceed at once to identifying groups and sub-groups, agreeing on lead cases or samples. Most cases will fall into this category. But in large, generally pharmaceutical or other consumer cases, the problems are more complex. The court must be pro-active in considering how best to progress the action so that valid claims are included and weak or hopeless claims excluded. This involves considering whether claimants should be required to join a register: to 'opt-in', or whether unidentified potential claimants should be deemed to be included unless they 'opt-out' and the notice that should be given to potential claimants. It also involves consideration of the best way of providing an effective filter. I consider each of these below.

40.    There are, however, difficulties in relation to test cases. Firstly, both claimants and defendants need to accept that the case will be a test case in relation, say, to liability for all those claimants in the same position. It is therefore necessary to make express orders in advance of determination that parties will be bound by the results. Secondly, there are also problems if the test case turns out to be atypical - if it is disposed of on particular grounds or if the judgment is couched in such a way that it leaves undetermined the similar issues in other cases. It is therefore necessary for the difficulties of identifying cases as test or lead cases to be specifically addressed by the court at an early stage. Thirdly, the current rules on legal aid make it difficult to pursue the test case approach, even though it may offer the most economical way of resolving actions affecting numbers of people in cases as diverse as pharmaceutical products and housing disrepair. It would clearly be sensible for the legal aid provisions to support, rather than to impede cost effective resolution by this means. I recommend that this should be looked at especially in relation to the sharing of costs between privately paying and legally aided clients.

41.    The Lloyds litigation gives an example of what can be achieved. At an early stage the court identified and decided a number of preliminary issues of principle common to one or more categories of cases. With the active co-operation of the Court of Appeal and the House of Lords, appeals were expedited. The court selected from cases in a particular category lead or pilot cases for trial as to liability and principles relating to quantum in the hope that decisions in these cases would provide firm guidance in relation to other cases in the same category.

'Opt-out' or 'opt-in'
42.    Typically multi-party rules in other jurisdictions adopt an 'opt-out' approach, in that a person's rights may be determined in a multi-party action without his or her express consent to or participation in the litigation (the approach is under rule 23 in the US, and similar rules in Ontario, British Columbia and Australia). Members of the group may, however, opt out - in other words, indicate that they wish to be excluded. If they opt out, a person is not able to benefit from any award of damages, although they may always bring a separate action. It has generally been considered that there would be difficulties in this jurisdiction in taking forward cases on an 'opt-out' basis because of the cost sharing rules, but the experience of 'opt-in' registers with cut-off dates has not been altogether positive or, indeed, helpful in resolving the allocation of costs, particularly since most multi-party actions are legally aided.

43.    For personal injury claims, it has been argued that an 'opt-out' scheme is unfair to defendants because it does not enable them to know the size of the group and the number of claims and their nature. The Law Society's working party therefore recommended an 'opt-in' approach with the establishment of a register at the initial stage of certification; provisions for varying the criteria for joining the register, as the case developed; and provisions for establishing cut-off dates and for costs sharing. This is the preferred approach where there is a well defined or identifiable group of claimants.

44.    There are, however, problems in establishing an 'opt-in' register too early in the life of a potential multi-party action where there is a large pool of unidentified claimants. Although the register may appear to give defendants an idea of the size of the group, experience has shown that early cut-off dates tend to result in a rush to register which encourages many weak or hopeless claims to be registered and inflates the pool of potential claimants. The bandwagon effect may raise unrealistic hopes of compensation from claimants. Adverse publicity may have a severe negative impact on the business of defendants at a stage when there has been insufficient investigative work to establish clear criteria for the claims or, in some cases, to establish any clear indication of causation.

45.    In some circumstances defendants and the Legal Aid Board may be well aware that there are large numbers of people who might be affected by the product in question. In those circumstances the claim may be more manageable if the initial certification puts any further individual applications for legal aid on hold and provides for deemed inclusion of unidentified potential claimants on an 'opt-out' basis until definitive criteria can be established to provide for the effective filtering of potential claims before they are entered on the register. There is, however, a need for action to be taken in relation to the limitation period and this can only be effective if there are provisions to suspend or freeze the running of the limitation period on certification of the MPS, as in many other jurisdictions, so that further claimants whose claims were not being considered in detail at this stage were not disadvantaged. This will require primary legislation. In the absence of such legislation I have no doubt that courts will continue to exercise their discretion to admit latecomers since the existence of the MPS ensures that defendants are already aware of the potential claims against them.

46.    The court should have powers to progress the MPS on either an 'opt-out' or an 'opt-in' basis, whichever is most appropriate to the particular circumstances and whichever contributes best to the overall disposition of the case. In some circumstances it will be appropriate to commence an MPS on an 'opt-out' basis and to establish an 'opt-in' register at a later stage.

Notice
47.    If members of a group are to opt out, or to join the register, they must know about the multi-party action. Notice may also be necessary at various other times throughout the course of the proceedings, eg, determination of generic issues; on settlement. In reaching the decision on notice the court must have in mind the cost of such notice and its usefulness: in some cases notice may be so expensive as to be disproportionate to the costs and benefits of the litigation, or it may not serve a useful purpose.

48.    In a multi-party action where there are many claims, each of which is small, there is little to recommend in a rule making notice to each potential claimant mandatory. The costs of identifying potential claimants, and preparing and sending the notice, will make the litigation as a whole uneconomic. In any event, where such claimants receive the notice and choose to opt out, they will receive nothing. Because with small claims it is uneconomic for them to litigate individually, they will almost invariably remain members of the group. In the United States, in small claims group actions, very few of the tens of thousands - in some cases millions - of potential claimants actually notified choose to opt out. Accordingly, courts must have the discretion to dispense with notice enabling parties to opt out having regard to factors such as the cost, the nature of the relief, the size of individual claims, the number of members of a group, the chances that members will wish to opt out and so on.

49.    Once the claims become more substantial, however, individual notice is economically possible. It is difficult to set a figure and the matter must be left to judicial discretion, taking into account the factors I have already mentioned. Yet even if the court decides that notice must be given to members of a group, it should have a discretion as to how this is to be done - individual notification, advertising, media broadcast, notification to a sample group, or a combination of means, or different means for different members of the group. In each case the court must take into account the likely cost and benefit before deciding on the course of action.

50.    The court should have a discretion to order by whom the advertising should be undertaken. The Law Society's working party recommended that the Law Society should provide further guidance to solicitors on advertisements placed in the early stages prior to the establishment of a group action. I welcome that. The Law Society also recommended that the timing and placement of subsequent advertisements should be approved by the court. There is also a need to approve the content of the advertisements and for the court to decide on the appropriate body who should place the advertisement - either the lead solicitor, the Law Society itself with its substantial media expertise, the Legal Aid Board or the court itself.

Establishing a filter
51.    It is important for the court to address one of the major problems identified in every response: the need to find better ways of weeding out weak and/or hopeless claims or, if possible, to prevent them entering the action in the first place. In legally aided cases, it also requires consideration of the best way in which the legal aid decisions on merits and cost-benefit can be meshed in with court procedures.

52.    Once sufficient investigative work has been completed, it should be possible to establish criteria for entry to or removal from the register. Such an agreement, reached at an early stage, even if that in itself takes several months to arrive at, should lead to a considerable saving of time and legal costs both to the Legal Aid Board and for the defendants. The court, in effect will have drawn up the criteria for the merits test that will be applied to each individual claim. It would also mean that parties who have no realistic chance of bringing a claim are weeded out at an earlier stage or not brought in. This will be beneficial not only for defendants but also for the individuals themselves and for those claimants with stronger claims who do proceed. It has been emphasised that the bandwagon effect, in cases such as benzodiazepine, has the effect of swamping stronger claims with a host of weaker claims, many of them with very questionable foundation, and making the action as a whole unviable. While criteria for entry will be of most concern in 'creeping disaster' cases including pharmaceutical and environmental claims, I see a clear need to establish equivalent criteria in all multi-party situations.

53.    Agreement on criteria should assist in drawing up standard questionnaires, agreed between the parties, the court and the Legal Aid Board. These would ensure that the initial information obtained from potential claimants enables all concerned to make a clearer assessment of the number of claimants who might actually have a case. They would enable the Board to make a more accurate cost-benefit assessment, than it can at present. They would provide the criteria for the merits test for the initial wave of entrants to a register if the Board decides to grant legal aid.

54.    In legally aided cases, the present arrangements for assessing the merits of potential multi-party actions rest largely with the Legal Aid Board. Many commentators consider that it has proved difficult to establish appropriate and satisfactory arrangements despite repeated endeavours. Suggested refinements include the obtaining of an independent opinion from counsel on the merits and allowing representations from defendants as in Scotland.

55.    Through the registration of an MPS the court will be involved from an early stage and will determine the shape and progress of the action. It will provide an independent focus on the preliminary investigative effort and will provide a more natural context in which to consider the defendant's representations. The Legal Aid Board has difficulty, at least procedurally, in dealing with these and getting the claimants' response to the defendants' allegations. It is difficult for the Board to 'adjudicate' between them. This essentially adversarial process is more naturally controlled by the court. The alternative is parallel assessment of the merits by the court and the Board. The preferred approach is for the court to delineate the shape of the action and determine the criteria which must be met by those wishing to join the action. The Board would make its decisions on funding in the light of the court's decision.

Costs of multi-party actions
56.    The Chief Taxing Master drew attention, at the Inquiry's seminar, to the need for the court to address the question of costs at an early stage, and for the judge to make costs sharing orders in respect of both claimants and defendants. These orders have to apply both to the costs of clients in respect of their own solicitors and of the opposing party should it obtain an order for costs. Orders on costs may need the assistance of the Taxing Masters if appropriate.

57.    If the treatment of costs is not examined from the outset, the result is either subsidiary litigation or protracted problems when the matter comes to taxation. My general proposals for information on costs to be made available at every stage when the managing judge is involved are all the more important in relation to multi-party actions, where many claimants will be legally aided and have no direct control over costs and where costs can escalate dramatically. At every stage in the management of the MPS the judge should consider, with the help of the parties, the potential impact on costs of the directions that are contemplated, and whether these are justified in relation to what is at issue. Parties and their legal representatives, as in other cases on the multi-track, should provide information on costs already incurred and be prepared to estimate the cost of proposed further work. It has been suggested that such examination should occur at intervals of three months. That must be for the managing judge to determine in each individual case.

58.    Other common law jurisdictions with a cost-shifting rule have not changed it when introducing special rules for multi-party actions. Multi-party actions are not so significantly different from ordinary litigation as to justify such a change. However, there are several respects in which the ordinary approach to costs needs to be modified.

59.    The court needs a wide discretion in deciding what are costs for the purposes of the ordinary rule. Multi-party actions involve costs which do not normally arise in individual litigation, such as co-ordinating and communicating within the group and liaising with the media in what are often high profile cases. At present the costs of action groups or claimants' co-ordinating committees are not generally met on taxation, although they have been in some cases. Yet without such groups it may be difficult to co-ordinate an action. It is also probably the case that effective co-ordination of the action saves defendants costs overall. It is necessary that the costs of action groups should be met on taxation and that a reasonable basis for acknowledging these, and any others considered necessary or appropriate, should be established and applied by the court from the start. The Law Society's working party recommended that work done to co-ordinate between claimants and their solicitors should be recoverable inter partes and I support this.

60.    Thirdly, in some cases it will be fair that the group as a whole bears a proportionate share of any costs. This was the approach in the well known decision involving the drug Opren, Davies (Joseph Owen) v Eli Lilly & Co. [1987] 1 WLR 1136, CA. However, this approach cannot be adopted as an invariable rule for multi-party litigation. The Scottish Law Commission has cogently set out the reasons (op.cit., p.278):

61.    The result of always adopting the Davies approach would be that in non-legally aided cases there would be a denial of access to justice. Indeed, this was evident in the Opren litigation itself. The claimants' legal advisers hoped that the lead cases could be chosen entirely from those legally aided claimants with nil contributions. When the Court of Appeal in Davies held that costs should be borne equally amongst all claimants, the privately funded clients were advised to discontinue because of the threat which costs posed to them. Ultimately a private benefactor agreed to underwrite their costs and the case was soon after settled without trial.

62.    It is therefore essential that the court approves any cost sharing arrangement at the outset, and that this includes any arrangements between the privately paying and legally aided claimants. Information on costs already incurred and to be incurred in the future will allow claimants to assess their eventual liability as the case develops.

Legal aid funding
63.    At present in most group actions the Legal Aid Board is underwriting the majority of claimants' costs. The Lloyds litigation, although it is largely privately funded, includes a number of legally aided claimants. In other smaller cases, notably housing cases, there is often a mix of legally aided and private claimants. Until now the cost of large product liability actions has been a significant deterrent to unassisted claimants. It therefore seems sensible to consider whether it would be possible for future actions affecting substantial numbers of people to be handled in a way which either combines funding from legal aid and private sources or extends financial eligibility to those who might normally be ineligible.

64.    Tight control by the court over the management of the case should reduce cost and make it more predictable. The arrangements I outline for establishing an effective filter prior to entry on the register should reduce the overall numbers of claimants and particularly those with weak claims. This of itself should enable privately paying litigants to enter into cases with more confidence than they can at present where costs are totally uncontrolled.

65.    Legally aided clients already contribute to the costs of their cases through contributions and the operation of the statutory charge. At present this is an open-ended commitment. I consider it appropriate that legally aided clients should continue to contribute to the costs, but there should be scope within a more managed system for this to be estimated as the case develops. There may also be scope to extend the upper limits of financial eligibility on the basis of increased contributions. In appropriate cases, with tight judicial management and control on costs it may be possible to make an estimate of overall liability in advance. Such an approach could be structured in a way which included a requirement to make a personal financial commitment to the action at the stage of initial entry to the register, and perhaps at later stages, when, for example, the judge imposes a cut-off date. At each stage the commitment should be for a fixed and finite amount rather than the present general open-ended liability.

66.    The requirement of a personal financial commitment would reduce the element of speculative litigation which is one of defendants' main concerns. If it could be balanced by a limit on individual claimants' liability for costs at each stage, that would meet the main concern of claimants.

67.    A number of those who represent the interests of claimants and consumers have suggested that such a scheme could be combined with the often-floated idea of a contingency legal aid fund (CLAF) funded by percentage success fees from successful claimants or an institutionalised conditional fee scheme for multi-party actions. They believe that their clients would be more than happy to forgo such a percentage if it meant that they could be on the register for a specified entry fee and that their maximum liability for costs could be known in advance. There may be interest in developing this approach from those currently providing legal expenses insurance. While personal injury cases could be funded on the basis of conditional fees without the establishment of a CLAF, such funding is a possibility at present for other multi-party actions. I hope that the Lord Chancellor's Department and Legal Aid Board can reconsider the possibility of a CLAF in the context of the greater financial control over litigation which my proposals represent.

68.    A precedent for such a multi-party action fund, started initially by government, but subsequently funded by a percentage levy on successful litigants, is Ontario's 'class proceedings fund', funded by the Law Foundation of Canada. With its counterparts elsewhere, the Law Foundation derives its income mainly from interest on trust accounts. The fund is to provide financial support for claimants to class proceedings in respect of disbursements - not legal costs generally - related to the proceedings, and for payments to defendants in respect of costs awards made in their favour against claimants who have received financial support from the fund. The principal advantages of such a fund are that it would be entitled to assist all multi-party litigants, not just those with incomes low enough to qualify for legal aid.

Protecting the overall interests of litigants
69.    The rationale behind multi-party actions is that the diminution of the individual rights of claimants and defendants makes the overall action more practicable and less costly to progress. But there is a need to ensure that those rights are protected: for defendants by the perceived fairness of the balance between generic issues and by establishing effective criteria for entry to the action. For claimants, the court has a more explicit role in ensuring that their interests are protected:

(a) in supervising the activity of lawyers;

(b) in ensuring the effective representation of their interests through the appointment of a trustee in appropriate cases;

(c) in approving settlement.

Lawyers and multi-party actions
70.    There is nothing wrong with lawyers taking the initiative in actions multi-party actions. A typical claimant in such cases is often poorly informed, or ignorant of the particular facts, and it will only be the lawyer who recognises the potential for claiming. Moreover, even if a claimant does suspect a violation of the law and seeks redress, the cost of doing so may act as a disincentive to action. Enhancing access to justice demands that those ignorant of their legal rights, or unable because of the cost to pursue them, be given the opportunity of vindicating them. If this requires lawyer initiative, then so be it.

71.    But because the lawyers will often be taking the initiative in multi-party actions, there are potential conflicts between their interests and those of group members. This can derive from the very reasons which make multi-party litigation attractive in the first place - the possible ignorance of potential claimants, and that they are disorganised and possibly also dispersed. Thus the opportunities for self-interested behaviour are generally greater in group litigation than in ordinary litigation. Particular forms which this has taken include bringing claims known to be unfounded for harassing purposes and genuine but limited value claims, knowing in both cases that defendants will feel impelled to settle on terms advantageous to the lawyer though possibly of little benefit to the group members. It would be remiss of the Inquiry not to make some recommendations to anticipate problems which experience here and elsewhere demonstrates can arise.

72.    In general terms the problems arise because of the relative absence of client control. When a group is large, members may not even be aware of the litigation until it is well under way. Even if they are aware of the litigation, how are claimants to have an influence? The view of individual claimants is greatly diluted, if not excluded, in a large group. As for a majority view, the costs of communicating between claimants, and organising meetings, may be so great as to make it impractical in many cases. If claimants are not involved in the conduct of litigation, however, they cannot really act as a monitor on the way the lawyers handle it.

73.    Among the strongest disincentives to meritless or frivolous multi-party litigation will be prompt dismissal by the courts. Court control from the very early days will ensure this. So too will an early determination of the merits. Courts must also be prepared to visit sanctions on lawyers who do not live up to the standards of professional behaviour expected. The Bar and the Law Society must give special attention to the ethical problems involved in multi-party litigation.

74.    Lawyers conducting multi-party litigation are entitled, of course, to reasonable remuneration but there are reports that working excessive hours and inflation of the time spent on a case are common abuses in multi-party action litigation in the United States. Where multi-party litigation in this country is legally aided, the Legal Aid Board has a duty to oversee the lawyers and to call a halt to this type of behaviour. Courts, too, have a role in this regard. I am recommending generally that costs should be actively considered by the judge throughout the case and that, if appropriate, a Taxing Master should also be involved throughout. Because of this continuing involvement, they will have a store of knowledge about the case. That involvement at the taxing stage will be invaluable. Moreover, if the lawyers know that the judge and his team managing the case may have an influence on their remuneration, this is likely to act as a strong incentive to proper and reasonable behaviour on their part.

Multi-party litigants and their support
75.    Multi-party actions are in any event an area of litigation which is even more lawyer-driven than any other. This may be exacerbated when the lead firm is funded on contract by the Legal Aid Board and the case will become driven by the legal team in conjunction with the Board. In those circumstances the court has a duty to ensure that the interests of the client group are protected. In the past it has been assumed that this was achieved by individuals being represented by their own solicitor and the action co-ordinated by a steering committee or lead firm. While this may provide a degree of local hand-holding and support, it does not enable the client group, either individually or as a whole, to assume the role of an informed client.

76.    But there is a wider issue here which may be particularly relevant in legally aided cases. That is the need to represent the interests of all the group, including those not specifically identified, and to ensure effective conduct of the litigation from the claimants' point of view.

77.    The Inquiry has heard how action groups can take on the role of an informed client, with formal constitutions established at the outset to provide for later problems, particularly in relation to settlement. Such groups can take account of their members' interests and ensure that these are reflected in the instructions to their legal representatives. Where there is no formal group representing the interests of the claimants, or where it is considered that the litigants' interests require separate representation, a trustee should be appointed by the court. There may also be a need for a trustee in cases where there are both privately paying and legally aided litigants, to ensure that the interests of both are taken into account. The trustee would be publicly funded, in some cases by the Legal Aid Board, on the basis that he or she would be fulfilling a role that would otherwise be met by an assisted person's own solicitors, or by arrangements under an 'all work' contract, which would require the lead firm to make arrangements for looking after individual clients as well as fulfilling a wider role.

78.    The role of trustee would be flexible but the main elements might be:

(a) to identify the objectives and priorities of the parties (by meeting them at an early stage to determine their needs), and to assist with devising a plan to meet those objectives;

(b) to maintain a watching brief on the public interest elements of the action to ensure that opportunities to instigate change are not missed;

(c) where necessary, to look after the interests of unidentified or unborn claimants and to act as protection against defendants picking out lead cases for settlement;

(d) if appropriate, to assist in the formation of an informal support group, if one does not come into being spontaneously (this could be done by advertising and holding regional meetings to inform people of the impending action and put them in touch with one another).

Approving settlement
79.    There is a strong case for court approval of all multi-party settlements, especially where the defendant offers a lump sum settlement, because:

(a) it is necessary to ensure that the lawyers do not benefit themselves while obtaining minimal benefit for their clients, or, alternatively, profiting from the vulnerability of commercially sensitive defendants;

(b) all members of the group are bound although they may be only indirectly represented;

(c) a lump sum settlement must be fair although it explicitly does not try to match individual loss exactly.

80.    There are two possible approaches to enabling the court to provide additional safeguards in this context. First, and particularly in cases where there may be unidentified or unborn potential claimants, the judge should satisfy himself that proper arrangements have been made, or request the trustee to do so.

81.    Secondly, the court could require an identified and finite group of claimants to have in place from the outset a constitution including provisions relating to acceptance of settlement, such as majority voting. The Inquiry heard how important this was considered to be by the action groups in the Lloyds litigation, which represented numbers of largely privately paying claimants. In such a case where a minority objected to the settlement, it would be open to the judge to hear their objections. The court may also have a role in administering settlements or resolving points of difficulty in borderline cases where criteria for settlement have been agreed.

82.    Experience elsewhere suggests that the court also has a role in cases which are before the court solely for settlement purposes. Experience, particularly in the USA, suggests that judicial oversight of settlements is not effective unless there are understood criteria for approval, which provide for cases which may be before the court solely for settlement purposes. Although the MPS is primarily a vehicle for managing actions, it could, if necessary, be requested to provide court oversight and approval of settlement. In such a case the criteria might cover such matters as whether:

(a) the pre-requisites for a multi-party situation have been met;

(b) the multi-party definition is appropriate and fair, taking into account, among other things, whether it is consistent with the purpose for which it is certified, whether it may be over-inclusive or under-inclusive, and whether division into sub-groups may be necessary or advisable;

(c) persons with similar claims will receive similar treatment, taking into account any differences in treatment between present and future claimants;

(d) notice to members of the group is adequate, taking into account the ability of persons to understand the notice and its significance to them;

(e) the representation of members of the group is adequate, taking into account the possibility of conflicts of interest in the representation of persons whose claims differ in material respects from those of other claimants;

(f) 'opt-out' rights are adequate to fairly protect interests of group members;

(g) provisions for lawyers' fees are reasonable, taking into account the value and amount of services rendered and the risks assumed;

(h) the settlement will have significant effects on parties in other actions pending;

(i) the settlement will have significant effects on potential claims of group members for injury or loss arising out of the same or related occurrences but excluded from the settlement;

(j) the compensation for loss and damage provided by the settlement is reasonable, taking into account the balance of costs to defendant and benefits to class members; and

(k) any claims process under the settlement is likely to be fair and equitable in its operation.

A special tribunal?
83.    In the issues paper published in January 1996, views were sought on the merits of establishing a special tribunal to act as a substitute for proceedings in court as to liability. The response has been overwhelmingly negative. There is a general consensus that the courts are rapidly developing case management techniques that will be further assisted by my general proposals and that the substitution of an inquiry would necessitate greatly increased funding to allow for the representation of the interests of the parties. It is considered that there would be no benefits to the process or to the funding of multi-party actions.

84.    There was also concern about my proposal that, in appropriate circumstances, a judge in charge of a multi-party action should move into 'inquiry' mode. The powers which the new rules will give to judges to control and limit evidence will result in far greater judicial control over the pace, scope and ordering of litigation. At a time of significant change this in itself represents a major shift of responsibility towards a more pro-active judicial involvement. I see no need for any further rules in this respect.

Inquiries
85.    The Law Society's working party recommended that legal aid be extended to boards or inquiries and that the costs of such representation should in principle be recoverable in any subsequent group action. It also recommended that there should be a presumption that any findings of fact be binding on the parties to any subsequent proceedings and inquests if the presumption was agreed by the parties before the inquiry. My own preference would be for a prima facie assumption that the findings are correct.

86.    The Scottish Law Commission has pointed out the difficulty of devising a single set of proceedings to serve with fairness all the purposes envisaged: the fatal accident inquiry is concerned with establishing, in the public interest, the circumstances surrounding particular fatalities; the criminal trial is obviously concerned with commission of a criminal offence, and there are strong rules excluding certain evidence in relation to this; and civil proceedings are concerned with person's claims, mainly to damages (Scottish Law Commission, Multi-Party Actions, Discussion Paper No.98, pp.50-1). It is clear from responses to the Inquiry that there appears to be a considerable element of duplication in the current approach to the establishment of disaster inquiries, inquests and subsequent criminal and civil litigation. There are also useful lessons to be learned from the study of previous inquiries and subsequent litigation. It was not part of the remit of my Inquiry to investigate this area but, despite the difficulties identified by the Scottish Law Commission, I consider that this is an area which requires further work, in particular in relation to its potential to inform or in part replace litigation in appropriate cases.

Conclusion
87.    Although in the Inquiry's issues paper I encouraged consideration of more radical alternatives to the new rule proposed by the Law Society's working party, I have been persuaded by the strength of the response that such approaches are not yet necessary, given the continuing development of more effective ways of handling multi-party actions. I hope that my proposals of a multi-party situation, of pro-active judicial control and prospective arrangements as to how costs should be dealt with (although not the actual amounts involved) as well as the other recommendations in this chapter, will contribute to that process. It will be for the Lord Chancellor, if these proposals meet approval, to develop them in conjunction with the relevant interests.

Recommendations
My recommendations are as follows.

(1) Where proceedings will or may require collective treatment, parties or the Legal Aid Board should apply for a multi-party situation (MPS) to be established. This would suspend the operation of the Limitation Act. The court may also initiate an application. Within the MPS, part of the proceedings could be common to some or all of the claimants, and other parts could be limited to individual claimants.

(2) Individual claimants would be able to join the MPS at the application stage and subsequently by entering their names on an initial register.

(3) The court should certify an MPS if it is satisfied that the group or groups will be sufficiently large and homogeneous, and that the cases within the MPS will be more viable if there is a collective approach than if they are handled individually.

(4) Lower value or local cases should be dealt with locally at appropriate courts by either a High Court or Circuit judge.

(5) A managing judge should be appointed at or as soon as possible following certification and should handle the action throughout.

(6) In appropriate cases additional support may be provided by the appointment of a deputy Master or deputy district judge from those practitioners who already have considerable experience of multi- party litigation.

(7) The court should have a residual power to approve the lead lawyer if a difficulty arises in appointing one.

(8) The court should usually aim to treat as a priority the determination of the generic issues while establishing economic methods of handling the individual cases.

(9) The court should have power to progress the MPS on an 'opt-out' or 'opt-in' basis, whichever contributes best to the effective and efficient disposition of the case.

(10) In reaching a decision on notice of the action to potential claimants, the court must take into account the cost of such notice and its usefulness.

(11) The court should be responsible for determining whether the action has merit and should proceed and the criteria which must be met by those wishing to join the action.

(12) The court should determine the arrangements for costs and cost sharing at the outset. The costs of action groups should be recoverable on taxation.

(13) The Lord Chancellor's Department and Legal Aid Board should consider the possibility of extending the upper limits of financial eligibility on the basis of increased contributions. In appropriate cases, with tight judicial management and control on costs it may be possible for assisted persons' liability to be assessed and fixed in advance.

(14) The possibility of a contingency legal aid fund should be reconsidered in the context of these proposals.

(15) The court has a duty to protect the interests of claimants, especially those unidentified or unborn.

(16) In appropriate cases the court should appoint a trustee.

(17) Multi-party settlements should be approved by the court especially where the defendant offers a lump sum settlement.

(18) The court should require an identified and finite group of claimants to have in place from the outset a constitution including provisions relating to acceptance of settlement.