1. INTRODUCTION

Interest in large-scale arbitration has grown exponentially in recent years, and the international legal community has seen the development of a variety of types of procedures. Class arbitration is perhaps the most well-known of these devices and is most closely associated with the United States due to the procedure’s resemblance to US-style class actions. 1However, two other countries — Colombia and Canada — have also considered adopting the procedure. 2The second large-scale device, collective arbitration, exists in a number of jurisdictions, including the United States, Spain and Germany. 3Ireland has also adopted a number of largescale collective procedures that can be considered arbitral or at least quasi-arbitral in nature. 4Although each country has developed its own unique procedural approach, all of these mechanisms involve individual as opposed to representative claims, as is the case in class arbitration. The final large-scale mechanism, mass arbitration, is seen exclusively in treatybased investment disputes and includes elements of both class and collective arbitration. 5

Some people attribute the development of large-scale arbitration exclusively to the US experience. 6However, other influences exist, most notably in the form of the increased attention being given to large-scale litigation around the world. 7Numerous countries, including those that once scorned the US class action as (yet) another form of American exceptionalism, have now developed their own means of providing largescale judicial relief. These mechanisms, often referred to as ‘collective redress’ so as to avoid confusion with US-style class actions, can provide useful models for new forms of large-scale arbitration. 8

At this point, the future of large-scale arbitration can best be described as mixed. On the one hand, class arbitration has been curtailed in the United States as a result of two US Supreme Court decisions upholding waivers of large-scale suits in arbitration. 9On the other hand, mass and multiparty arbitration appears to be on the rise in the investment context, with claims arising not only against Argentina but also against Cyprus. 10Collective arbitration appears to exist somewhere between these two extremes. While various procedures exist, parties do not appear to pursue these actions very frequently.

The fragmented nature of the legal environment involving large-scale arbitration gives rise to questions about the legal and social environments in which these procedures exist. In particular, doubts exist as to whether parties have sufficient incentives to pursue large-scale arbitration and whether states can or should do more to promote this particular form of dispute resolution. Questions also exist as to whether large-scale arbitration can arise in jurisdictions that do not provide for some type of large-scale legal relief in their national courts. 11

This chapter attempts to address these concerns by considering the issue of incentives for large-scale arbitration and in particular whether and to what extent policymakers can and should attempt to influence party behaviour through various mechanisms such as the use of default rules. 12Although questions relating to incentives and defaults have been considered in the context of large-scale judicial actions, 13this research has not yet been applied to arbitration. Furthermore, to the extent commentary regarding party motivation in large-scale arbitration does exist, it focuses on US-style class arbitration and does not consider other types of largescale proceedings.

The structure of this chapter is as follows. First, section 2 considers partyoriented incentives and disincentives to large-scale arbitration. Section 3 considers these issues from the perspective of the state to determine whether and to what extent a particular jurisdiction may want to either oppose or promote large-scale arbitration. Section 4 then considers what types of default rules and procedures might be used to generate an arbitral regime that best suits party and state interests. Section 5 contains various concluding remarks.

2. PARTY INCENTIVES AND DISINCENTIVES TO LARGE-SCALE ARBITRATION

The consensual nature of arbitration14means that parties will not adopt class, mass or collective proceedings unless they have sufficient incentives to do so. As a result, it is useful to consider the various reasons why parties might wish to proceed in large-scale arbitration as opposed to either large-scale litigation or bilateral arbitration. 15

Over the years, commentators have identified a number of reasons why parties might want to engage in large-scale proceedings. Although some of these rationales are somewhat limited in scope (in that they apply only to particular jurisdictions or only to one side of the dispute), it is still useful to consider them all so as to generate a comprehensive analytical framework.

A number of incentives for large-scale arbitration duplicate those seen in litigation. The most well-known of these involves efficiency, since a single large-scale proceeding can often be resolved in less time and for less money than a series of bilateral proceedings conducted either simultaneously (which would be difficult or impossible for both the respondent and any administering institution) or seriatim. 16Although this benefit inures most markedly to the respondent, there are also advantages to the claimants, since working together can reduce the time spent drafting pleadings and researching the law and the facts. Indeed, information-sharing among claimants has long been considered a benefit of large-scale litigation.

Joining multiple claims into a single forum also avoids the possibility of inconsistent determinations, something that respondents typically appreciate more than claimants. 17The concept of inconsistent determinations could arise on a cross-border basis (based on differences in the underlying law) 18or within single jurisdictions (based on differences in how the dispute is presented to the decision-maker). However, respondents may not view this feature in an entirely positive light, since the combination of claims into a single proceeding means that in the respondents lose, they lose big.

Commentators have long supported the use of large-scale dispute resolution devices based on the various benefits associated with a global settlement (ie, single resolution of all claims associated with a particular incident). To some extent, this perspective is justifiable, since respondents have been empirically shown to support global settlements. However, respondents only favour broad resolution of group claims after it becomes clear that the matter will be adjudicated on a large-scale basis. Up until that point, most respondents adopt a strategy that seeks either to have the claims thrown out of court or arbitration based on alleged ‘frivolity’19or to limit the number of persons who may assert a claim. Limitations can be substantive (as would be the case if a respondent were able to demonstrate that certain types of claims were legally insufficient) or procedural (as would be the case if a respondent were able to demonstrate a waiver of class or collective proceedings). In many cases, respondents believe (correctly) that their legal liability will be eliminated if they can eliminate the possibility of a group action, since many claimants will be unable or unwilling to proceed on a bilateral basis. 20

These types of defensive strategies are particularly predominant in the United States, where broad pre-trial discovery can sometimes allow claimants to obtain information about other victims or other possible types of wrongdoing and significantly expand the respondent’s legal exposure. Respondents in the United States are also eager to limit or avoid large-scale suits because the broad availability of punitive or treble damages make class and collective claims a threat to the respondent’s continued existence and viability. However, if a class proceeding cannot be avoided, US parties should want to be in arbitration rather than litigation, since the obligation to exchange information is far less onerous in arbitration than in litigation. 21Furthermore, arbitration does not use juries, which are often seen as one reason why class action judgements are so high. 22

Respondents in countries that do not contemplate the use of juries, broad pre-trial discovery or punitive damages may not see as many benefits to large-scale arbitration as respondents who may be sued in US state courts. However, parties in these jurisdictions may nevertheless identify other incentives (confidentiality, speed, cross-border coordination, etc.) involving large-scale arbitration.

One key incentive for large-scale arbitration that has not been considered in the literature involves the likelihood that most class, mass and collective suits will be resolved through settlement rather than adjudication. 23If this is indeed the case, then parties — particularly respondents — have a significant interest in being in arbitration, since the privacy associated with arbitration should help avoid the disclosure of information that would be embarrassing or legally problematic to the respondent. This is not to say that all types of large-scale arbitral proceedings can or should be entirely private and confidential (indeed, some arbitral institutions have already identified the various ways in which information about class, mass or collective proceedings should be made public), but the amount of publicity nevertheless appears to be more limited than in many types of litigation. 24

The likelihood that large-scale claims will settle before final resolution of the claims in arbitration or litigation identifies another reason why parties should prefer to initiate such matters in arbitration. At this point, it is very difficult to enforce a negotiated or mediated settlement agreement across national borders. While parties could seek to proceed under the Dutch Act on Collective Settlements (WCAM), which provides judicial confirmation to large-scale settlement agreements, 25not all parties will wish to invest the time and money associated with the Dutch proceeding. Instead, they may simply prefer to memorialize their agreement between themselves and hope that no dispute relating to the settlement agreement arises in the future. While there is currently no mechanism for enforcing such agreements in case of breach, 26some commentators have suggested that these types of settlement agreements could be deemed to be arbitral awards and enforced under the United Nations Convention on the Recognition and Enforcement of Foreign Arbitral Awards (New York Convention). 27This approach is as yet unproven, although commentators have suggested that it is important that any settlements that are intended to be enforced under the New York Convention be subject to a pre-existing arbitration agreement. 28

Although some parties, practitioners and commentators consider the above-mentioned incentives to large-scale arbitration significant, there are a number of disincentives to consider. The first and perhaps most significant concern for respondents is that by allowing large-scale arbitration, they open themselves up to larger numbers of claims and thus larger damages awards. 29This issue is perhaps most obvious in proceedings based on the Spanish statute for collective consumer arbitration, which requires notice of a large-scale proceeding to be given to the public at large. 30Similar concerns might arise in collective proceedings under the DIS Supplementary Rules for Corporate Law Disputes, although the notice given in those matters is much more limited than in the Spanish example. 31However, other types of actions — such as mass arbitration in the investment realm and class arbitration under the US model — do not involve a significant expansion of the number of claimants after the matter has been filed. In those cases, respondents attempt to have the entire matter deemed inadmissible, as reflected by battles over jurisdiction, admissibility and propriety of the underlying substantive claims.

As the preceding suggests, large-scale claims, however formulated, are often seen as ‘bet the company’ types of disputes. However, the nature of the issue is essentially the same in both litigation and arbitration, which means that large-scale arbitration cannot be considered inherently worse than large-scale litigation. Instead, the two procedures can be seen as largely analogous.

Although respondents are often the most vocal opponents to large-scale arbitration, claimants have their own sets of concerns. For example, claimants can often obtain a better outcome by proceeding individually than collectively, which should logically lead them to opt out or refuse to opt in to certain types of large-scale proceeding. 32However, many claimants fear the time, cost and energy associated with pursuing an individual litigation or arbitration and therefore prefer to proceed on a group basis.

Interestingly, one purported disincentive of arbitration involves the lack of a substantive appeal. This issue is raised most often by respondents, although defendants in large-scale judicial actions lose on appeal far more often than plaintiffs do, at least in the United States. 33As a result, this purported disincentive should actually be viewed by respondents as an incentive.

3. STATE INCENTIVES AND DISINCENTIVES TO LARGE-SCALE ARBITRATION

Although arbitration is traditionally framed as a party-driven procedure, states also have a significant interest in arbitral proceedings. 34Furthermore, states are capable of either encouraging or discouraging parties to engage in arbitration by altering the legal environment in which arbitration operates. These influences can be the result of judicial decisions (as has been the case in both Germany35and the United States36) or statutory enactments (as has been the case in Spain37and, to a lesser extent, Ireland38).

State incentives to class, mass, and collective arbitration can in some cases mirror incentives for similar types of judicial actions. For example, states would be just as concerned about inconsistent determinations in the arbitral context as in the judicial setting. However, some issues require a slightly different analysis due to the private nature of arbitration. For example, judicial efficiency is often touted as one of the primary motivations for both large-scale litigation and large-scale arbitration. However, the concept of judicial efficiency is slightly different in the arbitral setting because largescale arbitration does not consolidate numerous types of similar cases in a single court but instead removes those cases from the judicial system altogether. This particular feature might be extremely attractive in jurisdictions with extremely overburdened courts. Furthermore, as the Irish experience suggests, large-scale arbitration might be possible in counties that do not currently allow group actions in court, since the arbitration can sometimes provide for procedures that are not sanctioned in the judicial setting. 39

Arbitration might also be a way for states to make sure that large-scale disputes are heard by decision-makers with the necessary substantive and procedural expertise. To some extent, this conclusion may seem counterintuitive, since judges are often considered to be somehow superior to arbitrators. However, the evolution of international commercial arbitration suggests that states are willing to admit that arbitrators are at least as well-qualified as judges in matters involving cross-border commercial disputes. While it is unclear whether the confidence in arbitral competence relates to procedural matters associated with arbitration’s special blend of common law and civil law procedures or substantive concerns associated with arbitration’s frequent reliance on foreign law, the fact is that numerous states have promoted arbitration in international disputes, even in cases where domestic arbitration would not be appropriate. 40

Another potential incentive to large-scale arbitration involves the way in which class, mass and collective proceedings can be used as an ex post regulatory device. 41Using arbitration as a means of regulating certain behaviour may be particularly appropriate in international disputes, given the difficulties associated with traditional forms of international regulation. 42While some states may see this technique as inherently risky, the availability of the ‘second look’ doctrine would allow states to avoid enforcement of any award that violated a relevant public policy. 43

Although these incentives to allow large-scale arbitration are in many ways compelling, there are also a number of reasons why states may not wish to support class, mass, or collective arbitration. 44The first rationale involves the desire to avoid ‘secret’ proceedings on matters of wide social significance. Not only is arbitration seen as limiting the public debate about certain matters, but arbitration can also be seen as hindering the development of the law through judicial precedent. Confidential proceedings can also be considered as diminishing the deterrent value of large-scale actions, since other potential wrongdoers will not be aware of the consequences of improper behaviour.

States might also object to large-scale arbitration based on concerns about the parties’ procedural rights. Many of these concerns involve the use of representative mechanisms such as those seen in US-style class arbitrations. However, some types of mass and collective arbitrations can also give rise to potential problems not only with respect to the claimants’ ability to decide when, where, how and whether to assert a legal claim but also with respect to the ability of the respondent to defend against large-scale suits. 45Commentators have also raised concerns about collusion, particularly in the context of US class actions. 46Many of these issues have been addressed through various procedural protections, 47but in some cases the stigma associated with US-style class suits can lead states to oppose large-scale arbitration without undertaking a detailed analysis as to whether the procedure in question gives rise to the types of concerns at issue. Notably, not all concerns are relevant to all types of large-scale arbitration. 48

4. HOW STATES CAN AFFECT PARTY BEHAVIOUR

States can affect party behaviour in a variety of ways. For example, a state can adopt a particular law or regulation that explicitly prohibits or requires certain acts and can provide a penalty for non-compliance. However, this kind of direct approach may not be possible in cases where the state is not authorized to act. 49Thus, a law requiring a corporate entity to enter into a particular type of contract (such as an arbitration agreement) would be impermissible, since it exceeds the state’s ability to compel a particular action.

However, this is not to say that states cannot encourage parties to adopt certain behaviour that the state would be unable to control through direct regulation. Thus, a state can ‘encourage’ particular types of private activity by (1) creating a default regime that the parties would find difficult or expensive to change, even if change is technically or legally possible, (2) providing certain incentives to act in a particular manner, and/or (3) adopting certain disincentives to act in a particular manner. Each of these approaches can be used in the arbitral setting.

States that wish to undertake these types of actions can do so judicially or legislatively. At this point, the only legislature to have positively acted in favour of large-scale arbitration is Spain, which passed a statute providing for collective arbitration in consumer cases in 2008. 50However, at least one Canadian province has considered statutory action in the area of class arbitration51and Ireland has enacted a number of statutes allowing for various types of quasi-arbitral proceedings against the state. 52

Courts have also been involved in developing various forms of large-scale arbitration. 53The most notable example has been the United States, which has generated several types of class arbitration54as well as various forms of collective arbitration. 55Colombia has also adopted class arbitration by judicial decision. 56In 2009, the German Federal Court of Justice (Bundesgerichtshof or BGH) declared shareholder disputes arbitrable, thereby opening the door to large-scale arbitration of certain types of corporate disputes. 57Interestingly, the BGH only acted after the German legislature failed to address such issues following an invitation from the BGH in 1996 to consider such matters. 58

Before a state takes any direct or indirect steps to promote large-scale arbitration, it must decide that such a goal is desirable. Only after the determination to move forward with a particular policy has been made can the state identify the best means of achieving the enunciated goal. Each of these steps is considered below.

A. Step One
Determining Whether to Encourage Large-Scale Arbitration

When considering whether and to what extent to support large-scale arbitration, courts and legislatures must first determine whether there is a need for such a mechanism. 59This analysis is in many ways similar to the debate about the need for various forms of large-scale relief in the judicial context. Traditionally, common law jurisdictions (most notably the United States, Canada and Australia) considered large-scale litigation to be a necessary and appropriate means of supplementing the state’s regulatory scheme, 60while civil law jurisdictions (with a few exceptions, most notably Brazil) tended to view large-scale relief as both unnecessary and inappropriate. 61However, various changes in law and society have led numerous civil law jurisdictions to adopt various forms of collective redress. Thus, for example, more than half of the Member States of the European Union (EU) now offer some form of large-scale judicial relief as a domestic matter. 62Furthermore, the EU has identified a need to develop some form of cross-border collective relief so as to address the growing number of large-scale disputes involving claimants located in different countries. 63

These developments suggest a growing need for large-scale relief as a general concern. However, states must determine whether such relief can and should be offered in arbitration. In so doing, states must weigh the various incentives and disincentives identified in section two above.

The first thing to note when conducting this analysis is that a number of disincentives (namely, those relating to concerns about representative relief and those involving adequate protection of procedural rights) can be overcome through the adoption of proper procedures. 64Furthermore, some policy-related concerns about the use of large-scale arbitration (such as those relating to the need to develop precedent or the decreased ability to provide a deterrent effect) are inapplicable in those jurisdictions that do not use litigation as a means of developing law and policy or as a deterrent measure. 65

The diminished power of the various disincentives to large-scale arbitration suggests that states should be disposed to look favourably on the procedure. However, it is still necessary to consider the relative weight of the various state incentives to large-scale arbitration, since the mere lack of a disincentive is not enough to inspire a state to take a particular action. As it turns out, a number of the incentives discussed above (such as the ability of large-scale arbitration to reduce the burden on courts or regulatory enforcement mechanisms) are undeniable and can lead to significant public savings. 66While the strength of some potential incentives are subject to debate (for example, whether and to what extent arbitrators are better able than judges to manage large-scale disputes and whether large-scale arbitration can provide an equally or less expensive means of resolving mass disputes than litigation), many of the underlying concerns can be addressed through proper procedures (such as the requirement that at least one of the tribunal members be experienced in large-scale dispute resolution and specific provisions about cost-sharing among parties). 67

When considering these matters, states might find it useful to distinguish between domestic and international disputes, since large-scale arbitration may be particularly useful in cross-border matters. Thus, commentators have noted the way in which class, mass, and collective arbitration can overcome problems associated with (1) obtaining jurisdiction over all the parties, (2) determining and applying the governing law pursuant to a conflict of laws analysis, (3) harmonizing different national procedures and (4) facilitating international enforcement of the final decision. 68

Another issue that states may wish to consider when conducting this analysis involves the likelihood that most large-scale disputes are resolved by either a negotiated or mediated settlement. 69Although the Netherlands offers a relatively innovative means of enforcing international settlement agreements, the ability of a party to enforce other types of settlement agreements across national borders is very much in doubt. 70However, it has been suggested that parties to a pre-existing arbitration agreement may be able to have a settlement agreement declared to be a type of arbitral award that would be enforceable under the New York Convention. 71Thus, there may be benefits to having a pre-existing arbitration agreement in international disputes that are likely to settle. 72

Those states that decide to encourage large-scale arbitration can do so through a variety of means, as discussed in the next subsection. However, the situation does not require an all-in or all-out solution. Jurisdictions that are still undecided about the propriety of large-scale arbitration can adopt an intermediate step similar to that taken in Germany in cases involving corporate disputes. There, the BGH did not take affirmative steps to encourage collective arbitration but instead simply declared such matters arbitrable in theory and identified certain basic procedural protections that had to exist for a particular arbitration agreement or award to be considered valid. 73Thus, a court or legislature that wished to adopt a moderate approach to large-scale arbitration could declare such matters arbitrable (as Germany, Spain, Colombia and the United States have done, either explicitly or implicitly) and allow the precise procedures to evolve organically over time. Indeed, the task of creating detailed procedures for use in large-scale arbitrations might best be delegated to arbitral institutions that have the subject-matter expertise and the incentive to develop workable procedures.

While there is some risk associated with allowing arbitral institutions to develop large-scale procedures (since it might initially be unclear whether the arbitral institution has met all of the criteria considered necessary by the state), that task has been successfully carried out in both Germany and the United States. 74Furthermore, the current international arbitral regime evolved as a result of this type of trial and error process, so this approach is not without precedent. While it is impossible to discuss each of the relevant issues in the current chapter, states will likely pay particular attention to a variety of factors ranging from the nature of claimant participation (ie, whether parties proceed in a representative or aggregative (individual) capacity and whether participation proceeds on an opt-in or opt-out basis); how notice is to be provided to potential claimants; what alterations, if any, are to be made to the presumptions of privacy and confidentiality; how respondent consent may arise; and whether arbitral or judicial approval must be given to any settlements generated by the dispute.

Step Two
Determining How to Encourage Large-Scale Arbitration

At this point, several jurisdictions have demonstrated support for large-scale arbitration, although the level and type of support can vary both between different states and over time. When considering how best to encourage a particular procedure, states must consider whether to (1) create a default regime that the parties would find difficult or expensive to change, even if change is technically or legally possible; (2) provide incentives to adopt the procedure in question; and/or (3) adopt disincentives to adopting a procedure other than the one the state wishes to encourage. Each of these approaches is considered below.

4.B.1. Creating a Default Regime that the Parties Would Find Difficult or Expensive to Change

The legal literature is rife with information about how states can and do use default rules to affect party behaviour. 75Although these default rules are intended to maximize state interests, the rules do not operate as a traditional top-down regulatory mechanism. Instead, default rules operate in a coercive and often invisible manner to encourage parties to choose behaviours that the state wants to encourage.

Notably, the mere existence of legislation concerning large-scale arbitration does not necessarily constitute a default rule. Thus, for example, Spain has enacted a statute regarding collective consumer arbitration, but that provision does not operate as a default, since it does not drive the parties toward a particular procedure. 76Instead, the statute simply offers the parties an alternative means of resolving their disputes.

A different approach is reflected in a proposal that Gary Born made recently regarding a new international treaty that would make arbitration the default for all international commercial disputes. 77What makes this mechanism a default rule is the way in which the law requires parties to take certain affirmative steps to change the standard procedural regime. Rather than opting into a particular method of dispute resolution (as is the case with the Spanish statute), the parties would have to opt out of the proposed treaty regime. Since opting out requires some effort and incurs a certain amount of transaction costs, parties will be more likely to use the default mechanism, even if they might not otherwise choose to do so. 78

Default rules can often be recognized by the phrase ‘unless the parties agree otherwise’ and are extremely common in institutional rules on arbitration. 79Default rules on arbitration also exist, albeit to a more limited degree, in certain national laws80and international treaties. 81Thus, it would be possible for a state that wanted to encourage large-scale arbitration to do so through the creation of default rules. Indeed, one way to increase the use of the Spanish statute on collective arbitration would be to turn it into a default rule. 82Party autonomy would still be respected if the parties were allowed to opt out of the procedure, but the use of the procedure would likely increase if it were the default process.

These examples provide some basic illustrations of how a default rule relating to large-scale arbitration might be constructed. However, states only use default rules when they have a relatively high commitment to a certain mode of behaviour. At this point, it is unclear whether any particular jurisdiction has determined that large-scale arbitration is so useful that it should exist as the default in certain categories of cases.

This having been said, some states might be willing to consider a default rule on large-scale arbitration if they are presented with a concrete proposal on what that rule might entail. Unfortunately, that analysis cannot be conducted in the abstract. Instead, it is necessary to consider a variety of questions relating to the context and nature of the dispute (national or international), the mechanism in question (class, mass or collective), the relevant legal environment (common law or civil law) and whether the procedure is offered on a trans-substantive basis (as is the case with class and collective arbitration in the United States) or only with respect to certain subject matters (as is the case with collective arbitration in Spain and Germany and mass arbitration in the investment realm). It is also necessary to understand the legal and regulatory environment in which any proposed rule would operate, since any new procedure must work in tandem with existing procedures.

When thinking about whether to impose a default rule, states will likely be highly influenced by questions of timing, since default rules are typically imposed only in situations where the state has had some prior experience with a particular issue and can weigh the relative benefits of different approaches. Given the relative novelty of most forms of large-scale arbitration, it is unlikely at this point that states will adopt default rules encouraging class, mass or collective procedures. Instead, it is more likely that states will move gradually into this field and will begin by either increasing the incentives or decreasing the disincentives to group arbitration, since that approach allows the state to gather more evidence about the propriety of a particular rule or procedure before creating a default regime.

4.B.2. Providing Certain Incentives to Act in a Particular Manner

States often seek to affect party behaviour by providing incentives to act in a particular manner. Thus, for example, states wishing to minimize disputes about the service of process in litigation may give parties who waive formal process of service a longer time to respond to a complaint than parties who demand full and proper service.83 This approach encourages parties to avoid preliminary disputes about service without creating a default rule that could be seen as infringing on the parties’ procedural rights.

Incentives to party behaviour can be more difficult to identify than default rules. Indeed, some incentives have become so well-established that they are virtually invisible. However, such measures can nevertheless be found in the world of international commercial arbitration. Perhaps the most successful of these incentives involves the strong pro-enforcement policy reflected in the New York Convention. 84This policy has been one of the major reasons why parties have been attracted to arbitration over both litigation (which is the default regime) and mediation (which is another alternative mechanism that was more popular than arbitration in the years prior to the enactment of the New York Convention). 85States have consciously sought to incorporate this particular incentive into their national arbitration regime by joining the New York Convention. 86Use of incentives can also be seen on the domestic level in cases involving the United Nations Commission on International Trade Law (UNCITRAL) Model Law on International Commercial Arbitration (MAL). 87Although parties are still able and indeed entitled to seek judicial redress in international commercial disputes, adoption of the MAL makes arbitration so attractive that parties are likely to incur the necessary transaction costs (i.e., the cost of establishing an arbitration agreement) so that they can enjoy the benefits of this particular procedure.

One easy way for states to increase the incentives for large-scale arbitration would be to declare class, mass and collective procedures arbitrable as a matter of national law, thereby increasing the possibility that awards and agreements relating to such disputes will be deemed enforceable under the New York Convention. 88However, other types of incentives are also possible.

The most powerful measures would focus on expanding or solidifying existing reasons for parties to use large-scale arbitration. Thus, for example, actions that would decrease the costs associated with class, mass or collective dispute resolution would be effective in encouraging parties to adopt group procedures. These efforts could relate to measures to save litigation costs or increase the speed or finality of the process. Arbitration already offers some such measures (such as limits on the exchange of information and the ability to appeal on the merits), but additional actions might be adopted for large-scale disputes. For example, many respondents claim that the limited ability to have the claims dismissed during the preliminary stages of arbitration increases their transaction costs and results in more ‘strike suit’ settlements. 89Several arbitral institutions based in the United States have attempted to address that concern in class arbitration by creating two types of preliminary awards that could allow the dispute to be dismissed before the matter is heard on the merits. 90This system, which is somewhat similar to that adopted in investment proceedings, 91could be adapted for use in non-class proceedings.

Previously, this chapter suggested that respondents might be inclined to embrace large-scale arbitration as a means of avoiding inconsistent determinations, but only if having a single proceeding does not increase their financial exposure by bringing more claimants into the action than already exist or by allowing punitive damages. 92Thus, respondents might be encouraged to participate in large-scale arbitration if states found a way to limit the number of potential claimants93or eliminated the possibility of damages multipliers. While this latter option might not be much of an incentive in states that already limit recovery to consequential damages, there are enough countries in the world that allow punitive, exemplary or aggravated damages to make this proposal worth considering.

When adopting measure such as the elimination of damages multipliers, states must take care that they are not affecting the overall effectiveness of a particular system of justice. For example, commentators have noted that limiting the availability of punitive damages could be problematic in jurisdictions that typically fund large-scale dispute resolution through the use of contingency fees, since an attorney may not be willing to take a case where the final award is not calculated by reference to a multiplier. 94However, some of those difficulties might be addressed through the use of third party funders who may be willing to support a case even in the absence of punitive or exemplary damages. 95This is particularly true in the investment context, where the awards are often significant enough to generate the interest of third party funders even though damages are awarded on a compensatory basis. The same might be true of a large-scale competition law or cross-border consumer case. 96

Another potential incentive that the state might offer would be to allow parties in arbitration to use procedures that are not available in litigation. This measure might be particularly attractive in jurisdictions that have not yet adopted judicial forms of class or collective redress. Although some states might be concerned about the legitimacy of this approach, parties have long been allowed to consent to certain procedures in arbitration (such as fast-track or documents-only arbitration) that are not available in litigation. 97

States might also consider whether parties could be encouraged to use class, mass, or collective arbitration as a result of the principle of arbitral privacy and confidentiality. While many of the large-scale proceedings currently in place contemplate some derogation from the standard understanding of arbitral privacy and confidentiality, some degree of privacy or confidentiality nevertheless exists, thereby making arbitration more attractive than litigation. Certain additional provisions might also be put into place, such as measures prohibiting information that was disclosed in a large-scale arbitration from being used in other proceedings. 98

These are just a few of the possible ways to provide positive incentives to parties who might wish to engage in large-scale arbitration. Again, it is impossible to provide a comprehensive list of such measures in the abstract, since such recommendations must be made within the context of a particular legal environment. However, these types of policies could be combined with more general measures (such as the lengthening of time limits, as discussed in the context of the service of process) to encourage class, mass and collective procedures in appropriate circumstances.

4.B.3 Minimizing Any Disincentives to Act in a Particular Manner

States can also seek to affect party behaviour by minimizing any disincentives to act in a particular manner. Disincentives can be as difficult to identify as incentives, since they are often taken for granted in a wellestablished legal regime. However, it is possible to identify certain policy choices that eliminate or minimize disincentives that might otherwise limit parties’ willingness to use a particular procedure.

One example might be the way in which states allow arbitral institutions to require corporate entities to pay all or most of the fees and costs associated with consumer or employment arbitration. 99In so doing, the state helps remove one of the disincentives (cost) to claimants who might otherwise be inclined to enter into arbitration. While this measure could be seen as creating a disincentive to the respondent, most corporations prefer to have certain types of disputes (such as consumer, labour, employment and insurance matters) heard in arbitration for reasons ranging from savings of time and cost to avoidance of negative publicity and precedent. Therefore, the imposition of a small disincentive to the corporation is offset by the removal of a significant disincentive to the claimant.

Again, when considering how to encourage the use of large-scale arbitration, it is helpful to consider which disincentives are most problematic to the parties. The biggest concern appears to arise on the respondents’ side and is associated with the way in which group arbitration appears to expand the number of claims beyond that which would arise on a bilateral basis. Respondents have fought against this particular phenomenon by either requiring potential claimants to waive the right to proceed as a class or collective100or by declining to opt into systems that could result in an unknown number of claimants. 101The one type of large-scale arbitration that appears to avoid this problem involves corporate law disputes in Germany. However, the nature of those types of cases involves a relatively stable number of claimants, so the use of collective arbitration does not expand the respondent’s liability in any discernible way. 102

In many ways, questions of how to decrease disincentives are more difficult than those relating to increasing incentives, since a number of potential measures (such as reducing damages awards to eliminate respondents’ concerns about unlimited liability) could result in a violation of claimants’ procedural or substantive rights. Of course, this is not to say that solutions may not be found. For example, it might be possible to justify certain actions through analogies to other procedures (such as those relating to settlement agreements or insolvency actions, where creditors accept a percentage of what they would normally be entitled to receive in order to facilitate resolution of the dispute).

Other economically oriented solutions might look to the English Civil Procedure Rules for inspiration. Under those rules, a party may make a special type of settlement offer (a ‘Part 36’ settlement offer) to the court. 103If the offeree chooses not to accept a Part 36 offer, the offeree may continue to litigate the matter, although the offeree will be typically liable for the offeror’s litigation costs after the date of the settlement offer if the final judgment is less advantageous to the offeree than the proposed settlement. 104States might encourage both settlement and arbitration of large-scale disputes if they created a similar mechanism whereby the respondent was allowed to make a certified settlement offer to the arbitrator that capped damages at a particular sum. If the claimants chose to move forward with the arbitration, all costs after the date of the settlement offer would be borne by the claimants if the final award was greater than proposed settlement amount. This device may be particularly attractive in arbitration, where fees and costs can include not only attorneys’ fees but also arbitration fees and costs. The procedure encourages respondents to identify a reasonable amount for settlement, since an offer that is too low will easily be bettered after determination on the merits.

The other major disincentive to respondents is the relative difficulty of terminating an arbitration during the early stages of the dispute. However, that issue can be resolved by formalizing several intermediate stages where the matter might be dismissed, as discussed previously. 105

4.B.4 Balancing Measures

When considering various incentives and disincentives, it is important to consider how the proposed procedure works together as a whole. Commentators have noted that many judicial procedures offering collective redress are ineffective because they fail to provide parties on both sides of the dispute with sufficient rationales to use the mechanism. 106This problem usually arises because drafters are primarily concerned with avoiding the shortcomings of the US class action system. 107However, in many cases, reformers failed to appreciate how certain aspects of the US class action system made large-scale suits worthwhile to both plaintiffs and plaintiffs’ counsel. Without such incentives, neither victims nor lawyers may find it feasible to bring a large-scale suit. 108

States considering the possibility of large-scale arbitration should bear these issues in mind when seeking to create their own forms of class, mass, or collective arbitration. The system must be sufficiently even-handed to attract parties while also offering certain reasons to choose arbitration over litigation.

5. CONCLUSION

For years, the international community has wondered whether large-scale arbitration would become popular and what that procedure would look like. Up until this point, the development of large-scale arbitration has been delayed by two different obstacles.

First, lawyers and policymakers have been unsure as to whether there is actually a need for such mechanisms. Many commentators and policymakers took the view that ex ante regulation was in fact sufficient to control improper behaviour and provide redress to injured individuals. However, the evolution of judicial forms of class and collective relief around the world suggests that there is in fact a growing need to address largescale legal injuries. While it is possible that some jurisdictions’ domestic courts provide adequate protection for injured individuals, there may be times when arbitration offers a superior method of dispute resolution. One such example might be in cases involving cross-border disputes. In those situations, large-scale international arbitration offers the same significant benefits as international arbitration in bilateral proceedings. 109The recent increase in large-scale international disputes and the problems in developing inexpensive means of bilateral dispute resolution (such as through online dispute resolution (ODR)) 110suggests that new alternatives need to be developed.

The second obstacle to the development of large-scale arbitration involves the presumption that all such proceedings will reflect the norms used in US-style class arbitration. 111That presumption appears largely incorrect, given the criticism of both US-style class actions and US-style class arbitration. Furthermore, that presumption ignores the wide variation of collective redress procedures seen around the world and the growing number of mass and collective procedures in different jurisdictions.

At various times the preceding discussion has referred to the need to consider the precise shape of the proposed procedures before evaluating how to construct default rules and party incentives. Although this chapter does not take a view on what types of large-scale procedures are appropriate (since that question should be answered by reference to the particular type of legal injury the procedure is seeking to address and the legal environment in which the procedure is to operate), there are a variety of alternatives to consider as potential models: mass arbitration in the investment realm, collective consumer arbitration in Spain, collective corporate arbitration in Germany, and trans-substantive collective arbitration in the United States, just to name a few. 112Existing procedures can be blended, amended and transformed for use in other jurisdictions and other contexts. 113

Although states have an important role to play in the development of large-scale arbitration, it is unlikely that they will become involved at a detailed level. Instead, states will and likely should focus primarily on acts (legislative or judicial) that confirm the arbitrability of large-scale disputes and that identify any procedural minimums that must be met to allow such disputes to be arbitrated. 114Although it is possible for a state to participate in the creation of the precise procedures to be used, 115there are certain benefits to allowing the arbitral community to take the lead in developing procedures. Not only can arbitration experts be both innovative and efficient (as was the case in Germany, where the DIS created the Supplementary Rules for Corporate Law Disputes in a matter of months), but arbitral rules are more easily amended than laws, which allows the relevant procedures to evolve over time.

At this point, it may be useful for the arbitral community to seek to create international rules of procedure relating either to trans-substantive claims (ie, rules addressing any possible subject matter) or subject-specific claims (such as consumer, online or competition/antitrust disputes). Not only will such a process be useful to parties seeking to decide whether they wish to adopt such a procedure, it will also be useful to states seeking to determine whether and to what extent they wish to promote large-scale arbitration. Once a basic procedure is in place, the relevant stakeholders — states, parties, practitioners, institutions — can work to fine-tune the details over time, as has been the standard practice in both national and international arbitration.

The lead for this sort of initiative could come from a private institutional body with significant experience in both international and large-scale proceedings. While this recommendation may lead some people to think of the American Arbitration Association (AAA), which admittedly has a great deal of experience with class arbitration, 116that is a national institution that has close ties to a particular national procedure. Instead, a more international institution such as the International Chamber of Commerce (ICC) might be better placed to devise a truly international procedure that takes into account the predilections of both common law and civil law countries.

Public entities might also become involved in this discussion. While it is likely too early to ask UNCITRAL to develop a set of rules, since UNCITRAL works by international consensus and there is quite likely no consensus at the interstate level regarding the need for or form of international largescale arbitration, the Permanent Court of Arbitration (PCA) might be amenable to developing a set of procedures that could be used in international disputes. 117In many ways, the PCA appears to be an ideal choice, since the PCA not only has the trust of the international community but also has significant experience in administering mass claims through various claims tribunals. 118Indeed, the PCA has already assisted with the resolution of a large-scale international claim that was originally filed as a US class action. 119Although the PCA’s mass claims procedures are somewhat different from what is at issue in this discussion, in that the PCA procedures typically do not attempt to resolve all claims at a single time, in a single forum, there may nevertheless be some overlap between the two processes. 120Additional rules for large-scale matters might also be developed by the International Centre for Settlement of Investment Disputes (ICSID), particularly given the increasing incidence of such disputes in the investment context. 121Such an initiative would reduce the debate about the jurisdiction of the tribunal and the admissibility of the claims in ICSID proceedings and would therefore save the parties time and money.

Although the international legal and business communities may be concerned about the spread of large-scale arbitration, the situation is very different than it was in 2003, when the US Supreme Court decision in Bazzle v Green Tree Financial Corp shook the arbitral world to its very foundations. 122The development and expansion of large-scale arbitration has been gradual and well-considered, and there is every indication that future reforms will be taken with equal thoughtfulness. Furthermore, it seems likely that the fear and misapprehensions that often surround new procedural mechanisms are on the wane, and the national and international legal community can begin to discuss the future of class, mass and collective arbitration in a measured and thoughtful manner. It is hoped that this chapter has contributed to that discussion.

1
Strong, SI, Class, Mass, and Collective Arbitration in National and International Law (Oxford University Press 2012) 6-16, 38-42, 43-74 (hereinafter Strong, Class, Mass, and Collective Arbitration).

2
ibid 14-15 (noting Colombia has seen at least one class arbitration and Canada has considered the procedure both judicially and legislatively).

3
ibid 17-19, 36-37, 84-104; Strong, SI, "Collective Consumer Arbitration in Spain: A Civil Law Response to US-Style Class Arbitration," Journal of International Arbitration 30, 2013, 495 (hereinafter Strong, Spanish Statute) (discussing and translating Ley 231/2008); Strong, SI, "Collective Arbitration Under the DIS Supplementary Rules for Corporate Law Disputes: A European Form of Class Arbitration?" ASA Bulletin, 29, 2011, 45.

4
Strong, SI, "Large-Scale Dispute Resolution in Jurisdictions Without Judicial Class Actions: Learning From The Irish Experience," ILSA Journal of International & Comparative Law, 22, 2016, __ (hereinafter Strong, Irish). A number of matters have involved the state as respondent and thus could be considered administrative in nature. However, a recent large-scale product liability matter is entirely private and thus can be considered an arbitral procedure. ibid. See also "Go-Ahead for Alternative Dispute System Over Defective Implants: Move Aimed at Freeing Up High Court List and Ensuring Cases Against DePuy Decided Quickly," Irish Times (16 December 2015), available at <www.irishtimes.com/ news/crime-and-law/courts/high-court/go-ahead-for-alternative-disputesystem- over-defective-implants-1.2468833>.

5
Strong, Class, Mass, and Collective Arbitration (n 1) 16-17, 74-38. Investment arbitration has also seen a number of large multiparty proceedings that are somewhat akin to mass arbitration. Strong, SI, "Heir of Abaclat? Mass and Multiparty Proceedings: Ambiente Ufficio SpA v Argentine Republic," ICSID Review-Foreign Investment Law Journal, 29, 2014, 149; Strong, SI, "Consent in Multiparty Arbitration — The Most Recent Installment," (11 December 2014), <http://kluwerarbitrationblog.com>.

6
A number of US Supreme Court cases are considered central to this analysis. Keating v Superior Court, 645 P2d 1192 (Cal 1982), rev’d on other grounds sub nom Southland Corp v Keating, 465 US 1 (1984); Green Tree Financial Corp v Bazzle, 539 US 444 (2003); Stolt-Nielsen SA v AnimalFeeds Intl Corp, 559 US 662 (2010); AT&T Mobility LLC v Concepcion, 563 US 333 (2011); Oxford Health Plans LLC v Sutter, 133 SCt 2064 (2013); American Express Co v Italian Colors Restaurant, 133 SCt 2304 (2013).

7
Strong, Class, Mass, and Collective Arbitration (n 1) 2-3.

8
See Directorate General for Internal Policies, Overview of Existing Collective Redress Schemes in EU Member States, IP/A/IMCO/NT/2011-16, 5 (July 2011) (noting 16 of the then-27 Member States of the European Union had some form of collective redress, with more Member States contemplating development of large-scale judicial relief); Strong, Class, Mass, and Collective Arbitration (n 1) 282-83, 304-24.

9
AT&T Mobility LLC v Concepcion, 563 US 333 (2011); American Express Co v Italian Colors Restaurant, 133 SCt 2304 (2013). At this point, it is unclear whether and to what extent waivers of class claims extend to collective claims. Furthermore, there are significant policy issues associated with the ability of private parties to eliminate one of the primary regulatory mechanisms used in the US legal system. Strong, Class, Mass, and Collective Arbitration (n 1) 205-27 (discussing the role of private litigants as ‘private attorneys general’ in US law). Although the US law concerning waivers is problematic for parties in US class arbitrations, the US Supreme Court has indicated that class arbitration may proceed in cases where the arbitration agreement is silent or ambiguous as to class proceedings and that the arbitral tribunal is competent to decide that issue. Stolt-Nielsen SA v AnimalFeeds Intl Corp, 559 US 662 (2010); Oxford Health Plans LLC v Sutter, 133 SCt 2064 (2013); Strong, Class, Mass, and Collective Arbitration (n 1) 183-88. This point is occasionally misunderstood by commentators who mistakenly read Stolt-Nielsen as requiring explicit consent to class arbitration. ibid 183-84, 187 (noting Stolt-Nielsen allows implicit consent).

10
Strong, Class, Mass, and Collective Arbitration (n 1) 37 (discussing Abaclat v Argentine Republic, ICISD Case No ARB/07/5 (involving more than 60,000 claimants), Ambiente Ufficio v Argentine Republic, ICSID Case No ARB/08/9 (involving 90 claimants) and Alemanni v Argentine Republic, ICSID Case No ARB/07/8 (involving 74 claimants)); (n 5). Although large-scale investment arbitration begin with a trio of cases involving Argentina, a new series of disputes appears to have arisen involving Cyprus. Marfin Investment Group Holdings, SA v Republic of Cyprus, ICSID Case No ARB/13/27 (involving eighteen claimants); Adamakopoulos v Republic of Cyprus, ICSID Case No ARB/15/49 (involving 676 individual and institutional Greek investors with 434 claims); see also Stelios Orphanides, ‘Greek bailed-in investors and bondholders seek €120m in damages with ICSID appeal’ Cyprus Business Mail (29 September 2015), http://cyprusbusinessmail.com/?p=15139 (discussing a newly filed ICSID arbitration involving 676 claimants).

11
This question appears to be put to rest by developments in Ireland. See Strong, Irish (n 4).

12
The author has considered this issue to some extent previously. Strong, SI, "Class Arbitration Outside the United States: Reading the Tea Leaves," in Hanotiau, Bernard, and Schwartz, Eric A (eds), Dossier VII: Arbitration and Multiparty Contracts (ICC Institute of World Business Law 2010) 183, 194-206 (hereinafter Strong, Tea Leaves).

13
Strong, SI, "Regulatory Litigation in the European Union: Does the US Class Action Have a New Analogue?" Notre Dame Law Review, 88, 2012, 899 (hereinafter Strong, Regulatory Litigation); Hensler, Deborah R, "The Globalization of Class Actions: An Overview," in Hensler, Deborah, Hodges, Christopher and Tulibacka, Magdalena (eds), The Globalization of Class Actions, 622 The Annals of the American Academy of Political and Social Science (2009) 7.

14
Some authorities believe that consent to a large-scale proceeding can be implicit and arise as a result of the initial decision to arbitrate. Strong, Class, Mass, and Collective Arbitration (n 1) 172-80, 183-205. Other authorities take the view that parties must explicitly consent to a particular form of large-scale arbitration. ibid. It is impossible to resolve this particular debate within the context of the current discussion.

15
ibid 284-304 (discussing the relative merits of large-scale arbitration and large-scale litigation).

16
ibid 82 (discussing the preliminary award in Abaclat v Argentine Republic, which noted that ICSID would be unable to administer 60,000 individual claims simultaneously).

17
For example, Argentina has expressed concerns about inconsistent awards in investment arbitration on various occasions, which could be seen as undercutting its objection to mass proceedings. ibid 274.

18
Some concerns could arise if claimants who have experienced identical injuries are treated differently as a result of differences in the underlying law. ibid 287.

19
Empirical studies show that most class suits are not as frivolous as respondents believe. ibid 8, 67 (citing studies).

20
ibid 207, 343 n395.

21
Although the scope of disclosure in US-style class arbitration might be slightly less than the scope of discovery in US class actions, it is unlikely that an arbitral tribunal would prohibit the exchange of information in a class arbitration.

22
ibid 249, 295.

23
The conventional wisdom is that most large-scale actions settle prior to final resolution in court or in arbitration, since adjudicative mechanisms are both costly and risky. ibid 68 (discussing class arbitrations), 234 (discussing judicial class actions). See also Hensler (n 12) 19-20.

24
While some rules on large-scale arbitration provide for limited types of public disclosure (for example, class proceedings administered by the American Arbitration Association, collective proceedings administered by the Deutsche Institution für Schiedsgerichtsbarkeit (DIS) and mass proceedings administered by the International Centre for Settlement of Investment Disputes (ICSID) will be subject to a modified rule of confidentiality), other procedures do not (for example, class arbitrations administered by JAMS). Strong, Class, Mass, and Collective Arbitration (n 1) 64-66.

25
The WCAM is available on an opt-out basis for parties to both domestic and cross-border disputes. See Wet collectieve afwikkeling van massaschades (WCAM), Law of 23 June 2005, Stb 340; Tzankova, Ianika, and Lunsingh Scheurleer, Daan, "The Netherlands," in Hensler, Deborah, Hodges, Christopher, and Tulibacka, Magdalena (eds), The Globalization of Class Actions, 622 The Annals of the American Academy of Political and Social Science (2009) 149, 153-55. Notably, the WCAM does not require the assistance of a third party neutral. Instead, the parties may simply negotiate a settlement on their own.

26
A new international instrument is currently under discussion. Strong, SI, "Use and Perception of International Commercial Mediation and Conciliation: An Empirical Study" (forthcoming 2016) (hereinafter Strong, Empirical) (discussing efforts by the United Nations Commission on International Trade Law (UNCITRAL) to create a legal enforcement regime for mediated settlements which are currently quite difficult to enforce internationally).

27
Convention on the Recognition and Enforcement of Foreign Arbitral Awards, 10 June 1958, 330 UNTS 38 (hereinafter New York Convention); Strong, SI, "Beyond International Commercial Arbitration? The Promise of International Commercial Mediation," Washington University Journal of Law & Policy, 45, 2014, 11, 36 (hereinafter Strong, Mediation).

28
Newmark, Christopher and Hill, Richard, "Can A Mediated Settlement become an Enforceable Arbitration Award?" Arbitration International, 16, 2000, 81, 81-87; Strong, Mediation (n 27) 36. See also Strong, Class, Mass, and Collective Arbitration (n 1) 315-16.

29
Hensler (n 12) 25.

30
Strong, Spanish Statute (n 3) 501, 508-10 (discussing provisions requiring notice to be posted in the Official Journal).

31
Strong, Class, Mass, and Collective Arbitration (n 1) 94-96 (indicating that notice is generated as a result of information given by existing parties and involving ‘Concerned Others’).

32
US-style class actions tend to proceed on an opt-out basis, while various forms of collective and mass arbitration tend to proceed on an opt-in basis. ibid 7, 18-19, 78. However, it is unclear whether and to what extent the Spanish statute on collective consumer arbitration allows claimants to avoid group determination of their rights if the respondent chooses to proceed collectively. Strong, Spanish Statute (n 3) 504-05. That approach could be problematic from a constitutional or human rights perspective. ibid 505-06.

33
Strong, Class, Mass, and Collective Arbitration (n 1) 134.

34
Strong, SI, "Discovery Under 28 U.S.C. §1782: Distinguishing International Commercial Arbitration and International Investment Arbitration," Stanford Journal of Complex Litigation, 1, 2013, 295, 350-67 (discussing state interests in both private (contract-based) and public (treaty-based) arbitration).

35
In 2009, the German Federal Court of Justice (Bundesgerichtshof or BGH) created the possibility of large-scale arbitration in certain types of corporate disputes after the legislature failed to act. Strong, Class, Mass, and Collective Arbitration (n 1) 86.

36
In 2003, the US Supreme Court provided significant impetus to class arbitration with its decision in Green Tree Financial Corp v Bazzle, 539 US 444 (2003), although large-scale arbitration had existed in the United States for at least twenty years before that. Strong, Class, Mass, and Collective Arbitration (n 1) 7-12 (discussing Bazzle in addition to Keating v Superior Court, 645 P2d 1192 (Cal 1982), rev’d on other grounds sub nom Southland Corp v Keating, 465 US 1 (1984) and various state law and other developments). However, the Court reversed its support for class arbitration in 2009 and 2010 with a series of decisions curtailing class arbitration in some situations. Stolt-Nielsen SA v AnimalFeeds Intl Corp, 559 US 662 (2010); AT&T Mobility LLC v Concepcion, 563 US 333 (2011); Oxford Health Plans LLC v Sutter, 133 SCt 2064 (2013); American Express Co v Italian Colors Restaurant, 133 SCt 2304 (2013). See also (n 9) (discussing waiver and contractual silence). Although class arbitration still exists in the United States, it is nowhere near as prevalent as it was between 2003 and 2010. Strong, Class, Mass, and Collective Arbitration (n 1) 12.

37
Strong, Spanish Statute (n 3) 208-10 (translating Ley 231/2008).

38
Although Ireland has enacted a number of statutes requiring quasi-arbitral proceedings in situations where large-scale claims have been brought against the state, such statutes do not appear appropriate or possible in purely private actions. Strong, Irish (n 4).

39
Born, Gary B, International Commercial Arbitration (Kluwer Law International 2014) 1518 n593; Strong, Class, Mass, and Collective Arbitration (n 1) 304. See also Strong, Tea Leaves (n 12) 194-206. Indeed, this appears to be the case in Ireland, where a number of large-scale arbitrations or quasi-arbitrations appear to proceed on a documents-only basis, even though most if not all judicial proceedings contemplate an oral hearing.

40
Mitsubishi Motors Corp v Soler Chrysler-Plymouth, Inc, 473 US 614, 629 (1985).

41
Strong, Class, Mass, and Collective Arbitration (n 1) 2, 228-81.

42
ibid 235-41 (discussing various problems associated with transnational regulation, including the possibility of regulatory and procedural ‘mismatches’).

43
Mitsubishi Motors Corp v Soler Chrysler-Plymouth, Inc, 473 US 614, 638 (1985); Born (n 39) 980-81.

44
Of course, that ‘support’ could be exhibited at two different stages — when parties seek to enforce an arbitration agreement and when they seek to enforce an arbitration award.

45
Thus, for example, the Spanish statute on collective consumer arbitration may limit the ability of claimants to decide when, where, how and whether to assert a legal claim. Strong, Spanish Statute (n 3) 505-06.

46
For example, counsel for the plaintiff class may be tempted to collude with the defendant to obtain a fast and easy settlement that benefits the lawyers (who are often retained on a contingent fee basis) but that is not really in the best interest of the injured plaintiffs.

47
For example, in US-style class arbitrations, claimants can often opt out of actions. Furthermore, an arbitrator is supposed to review any settlement agreement so as to reduce the chance of collusion. Strong, Class, Mass, and Collective Arbitration (n 1) 322-23.

48
For example, concerns about collusion would likely not exist in cases where lawyers were not paid based on a percentage of the final award.

49
For example, states may not act in areas that are entirely private. Dorsett Niles, John, Tribble, Lauren E and Wimsatt, Jennifer N, "Making Sense of State Action," Santa Clara Law Review, 51, 2011, 885, 890; Strong, Irish (n 4).

50
Strong, Spanish Statute (n 3) 508-10 (translating Ley 231/2008).

51
Strong, Class, Mass, and Collective Arbitration (n 1) 15, 244-45. See also Strong, SI, "Resolving Mass Legal Disputes Through Class Arbitration: The United States and Canada Compared," North Carolina Journal of International Law & Commercial Regulation, 37, 2012, 921 (discussing Supreme Court and provincial cases in Canada). To date, one Canadian province, Manitoba, has considered this issue explicitly. Manitoba Law Reform Commission, Report 115, Mandatory Arbitration Clauses and Consumer Class Proceedings 3-4, 22-23 (April 2008). See also MacKinnon v National Money Mart Co, 2009 BCCA 103, ¶14 (calling for legislative consideration of class arbitration in British Columbia).

52
Strong, Irish (n 4).

53
Canadian courts have declined the opportunity to adopt large-scale arbitration as a common law device and have instead suggested that such issues are more properly subject to legislative action. Strong, Class, Mass, and Collective Arbitration (n 1) 15, 244-45.

54
The earliest type of class arbitration was developed in US state courts beginning in the 1980s. ibid 7-12 (discussing the effect of Keating v Superior Court, 645 P2d 1192 (Cal 1982), rev’d on other grounds sub nom Southland Corp v Keating, 465 US 1 (1984), as well as state law and other developments). The procedure evolved over the years into a more sophisticated system, particularly after the advent in 2003 of various institutional rules on class arbitration. ibid (discussing the effect of Green Tree Financial Corp v Bazzle, 539 US 444 (2003)).

55
Strong, Class, Mass, and Collective Arbitration (n 1) 84-85.

56
ibid 14-15.

57
Bundesgerichtshof 6 April 2009, II ZR 255/08 (Germany), digest by Kriendler for Institute for Transnational Arbitration (ITA), available at www.kluwerarbitration. com. See also Borris, Christian, "Arbitrability of Corporate Law Disputes in Germany," in Klaassen, CJM, and others (eds), Onderneming en ADR (Wolters Kluwer 2011) 55, 59-64. The DIS completed the process by adopting arbitral rules that complied with the BGH’s requirements for those types of disputes. Strong, Class, Mass, and Collective Arbitration (n 1) 86. While the current DIS rules refer only to disputes involving closely held corporations, the mechanism would appear equally appropriate in cases involving larger companies or multinational entities. ibid.

58
Strong, Class, Mass, and Collective Arbitration (n 1) 86.

59
Strong, Tea Leaves (n 12) 197-205.

60
Mulheron, Rachael, The Class Action in Common Law Legal Systems: A Comparative Perspective (Hart Publishing 2004). The United Kingdom has resisted large-scale litigation and has adopted a mechanism (the Group Litigation Order or GLO) that reflects aggregate as opposed to representative relief. Civil Procedure Rules (England), Rules 19.10 to 19.15; Practice Direction 19B — Group Litigation.

61
Gidi, Antonio, "Class Actions in Brazil — A Model for Civil Law Countries," American Journal of Comparative Law, 51, 2003, 311.

62
Directorate General for Internal Policies (n 8) 5.

63
Strong, Regulatory Litigation (n 12) 904-06 (discussing various EU initiatives). The EU is particularly concerned about large-scale disputes involving consumers and competition law.

64
Some examples of ‘proper procedures’ can be found in the mass and collective context.

65
ibid 907, 911 960-64 (noting European reliance on compensatory relief rather than regulatory or deterrent effects).

66
ibid 945-46 (discussing decreasing public budgets to deal with regulatory non-compliance).

67
Strong, Class, Mass, and Collective Arbitration (n 1) 67-68, 99-100 (discussing AAA and DIS approach to fees and costs in large-scale arbitration).

68
Strong, SI, "Class and Collective Relief in the Cross-Border Context: A Possible Role for the Permanent Court of Arbitration," The Hague Yearbook of International Law 2010, 23, 2011, 113, 124-33.

69
(n 23).

70
(n 25). At the very least, such efforts will be expensive and time-consuming. See also Strong, Mediation (n 27) 26-28, 36; Strong, Empirical (n 26).

71
Strong, Mediation (n 27) 36 (noting that mediated settlement agreements may be considered a type of arbitral award for purposes of the New York Convention, although that approach is disputed). Steps are being taken to increase the international enforceability of mediated settlements. Strong, Empirical (n 26).

72
While steps are being taken by the international community to eliminate the disparity between matters that are subject to an arbitration agreement and those that are not, it is unknown when the international community will act. See Report of Working Group II (Arbitration and Conciliation) on the work of its sixty-third session (Vienna, 7-11 September 2015), UN Doc A/CN.9/861 (17 September 2015) (noting work on an international instrument concerning international settlement agreements arising out of mediation); Strong, Empirical (n 26).

73
Strong, Class, Mass, and Collective Arbitration (n 1) 18-19.

74
This has been the case in both Germany and the United States. ibid 40-74, 86-101.

75
There is a considerable body of literature concerning default rules. Although interesting, this discussion is outside the scope of the current chapter. See eg Ayres, Ian, "Valuing Modern Contract Scholarship," Yale Law Journal, 112, 2001 881; Ayres, Ian and Gernter, Robert, "Filling Gaps in Incomplete Contracts: An Economic Theory of Default Rules," Yale Law Journal, 99, 1989, 87; Ayres, Ian and Gernter, Robert, "Majoritarian vs. Minoritarian Defaults," Stanford Law Review, 51, 1999 1591; Ayres, Ian and Gernter, Robert, "Strategic Contractual Inefficiency and the Optimal Choice of Legal Rules," Yale Law Journal, 101, 1992, 729; Johnson, Jason Scott, "Strategic Bargaining and the Economic Theory of Contract Default Rules," Yale Law Journal, 100, 1990, 615; Posner, Eric A, "Economic Analysis of Contract Law After Three Decades: Success or Failure?" Yale Law Journal, 112, 2003, 829. However, it is clear that the state’s choice of a default rule will affect the bargaining positions of the parties. Gidi (n 61) 338; Rau, Alan Scott and Sherman, Edward F, "Tradition and Innovation in International Arbitration Procedure," Texas International Law Journal, 30, 1995, 89, 117 n152.

76
Furthermore, having a law, of itself, does not appear to be sufficient to have encouraged parties to use the procedure. See Strong, Spanish Statute (n 3) 496-97.

77
Perry, Sebastian," BITS, BATS and Buts," Global Arbitration Review (28 January 2013). See also Draft Model Bilateral Arbitration Treaty, https://www.wilmerhale. com/uploadedFiles/Shared_Content/Editorial/News/Documents/Draft-Model- BAT.pdf; Gary Born, BITS, BATS, and Buts: Reflections on International Dispute Resolution (speech at the University of Pennsylvania) https://www.wilmerhale. com/uploadedFiles/Shared_Content/Editorial/News/Documents/BITs-BATsand- Buts.pdf.

78
Rau and Sherman (n 75) 117 n152 (noting parties may be ‘reluctant to opt out even of rules that are inefficient for them’ if the transaction costs to alter the default rule are too high or onerous).

79
Any provision that reads ‘unless the parties agree otherwise’ constitutes a default rule.

80
For example, most national arbitration laws indicate that the court at the seat will assist with the selection of the arbitrator(s) unless the parties agree otherwise. See eg 9 USC s 5; Arbitration Act 1996 s 16.

81
For example, the Inter-American Treaty on Recognition and Enforcement of Foreign Arbitral Awards (Panama Convention) contains a set of procedural rules that applies unless the parties agree otherwise. Inter-American Convention on International Commercial Arbitration, 30 January 1975, art 3, OASTS No 42, 14 ILM. 336 (1975).

82
To do so, the statute would have to be amended to require parties to use the collective procedure in question rather than to simply permit them to do so.

83
US Federal Rules of Civil Procedure 4(d)(3), 12.

84
Born (n 39) 232.

85
ibid 73; Strong, Mediation (n 27) 12 (noting mediation and conciliation were the preferred means of resolving international commercial disputes prior to World War II).

86
New York Convention (n 27).

87
UNCITRAL, UNCITRAL Model Law on International Commercial Arbitration, 18th Sess, Annex 1, UN Doc A/40/17 (21 June 1985), revised by UNCITRAL, Revised Articles of the UNCITRAL Model Law on International Commercial Arbitration, 39th Sess, Annex, UN Doc A/61/17 (7 July 2006).

88
Strong, Class, Mass, and Collective Arbitration (n 1) 163 (quoting Gary Born), 243-47 (discussing the concept of ‘procedural non-arbitrability’).

89
ibid 316-18, 322.

90
ibid 46-57 (discussing certain partial final awards built into the AAA and JAMS class arbitration regimes).

91
Investment tribunals often render preliminary awards on admissibility and jurisdiction.

92
Of course, some countries do not allow punitive damages to be recovered because of their deterrent or criminal law function. Born (n 39) 3077-82.

93
For example, use of an opt-in rather than opt-out system tends to decrease the number of claimants, since people must take an affirmative step to become involved in a particular dispute. However, respondents still may experience some concerns about how many claimants might join into an action if the respondents are required to give wide notice of the pending proceeding, as is the case under the Spanish statute. Strong, Spanish Statute (n 3) 501-03.

94
In some jurisdictions, most notably the United States, punitive or treble damages are used as an incentive to plaintiffs and plaintiffs’ counsel to bring cases that operate for the public good. Strong, Class, Mass, and Collective Arbitration (n 1) 259-60.

95
ibid 323, 353. See also Shannon, Victoria A, "Harmonizing Third-Party Litigation Funding Regulation," Cardozo Law Review, 36, 2015, 861, 865.

96
The economics and incentives associated with domestic competition or antitrust law disputes have been well-documented, as have the economics and incentives of investment arbitration. Strong, Class, Mass, and Collective Arbitration (n 1) 113, 343.

97
ibid 163 (quoting Gary Born as stating ‘it is difficult to see what . . . nonarbitrability objections could be raised to class arbitration’).

98
This rule already exists in some national courts. Civil Procedure Rules (England), Rule 31.22 (limiting subsequent use of documents or information produced in litigation).

99
See eg AAA, Consumer Arbitration Rules, 33-36 <www.adr.org>.

100
Strong, Class, Mass, and Collective Arbitration (n 1) 205-27.

101
Strong, Spanish Statute (n 3) 496-97.

102
Strong, Class, Mass, and Collective Arbitration (n 1) 19 (discussing how the collective arbitration rules apply only to closely held corporations). Although this procedure is relatively new, it has already enjoyed some success. ibid.

103
Civil Procedure Rules (England), Part 36; Practice Direction 36A — Offers to Settle.

104
The rules vary slightly depending on who made the offer and what type of dispute is at issue, but the basic principle holds true. Civil Procedure Rules (England), Part 36.

105
(n 90).

106
Strong, Regulatory Litigation (n 12) 960-61, 966-67 (discussing problems with European collective redress systems).

107
ibid (noting concerns regarding representative relief, broad discovery, punitive damages and contingent fees). Commentators have noted that many jurisdictions are hampered in their efforts to create a workable system of collective redress by the unavailability of contingent fees or third-party funding and the imposition of the loser-pays rule under national law. ibid.

108
ibid.

109
(n 68) (discussing jurisdiction, conflict of laws, procedure and enforcement).

110
At one time, ODR appeared to be on the fast track to standardization. Eidenmuller, Horst and Engel, Martin, "Against False Settlement: Designing Efficient Consumer Rights Enforcement Systems in Europe," Ohio State Journal on Dispute Resolution, 29, 2014, 261, 275-76 (discussing ODR initiatives from the EU and UNCITRAL). However, efforts at UNCITRAL have been delayed and are now moving forward in a much more fractured form. ibid. See also Annotated Provisional Agenda, UNCITRAL Working Group III (Online Dispute Resolution), at 2-4, UN Doc A/CN.9/WG.III/SP.135 (28 September 2015) (noting nearly fifteen-year history of the debate about ODR at UNCITRAL).

111
Billiet, Philippe, Class Arbitration in the European Union (Maklu Publishers 2013).

112
Strong, Class, Mass, and Collective Arbitration (n 1) 27-104. Inspiration might also come from procedures used by mass claims tribunals, although those processes typically do not attempt to resolve all disputes at a single time, in a single forum. ibid 330-32.

113
For example, the DIS Supplementary Rules for Corporate Law Disputes would be very useful as a means of resolving large-scale disputes involving trusts. Strong, SI, "Mandatory Arbitration of Internal Trust Disputes: Improving Arbitrability and Enforceability Through Proper Procedural Choices," Arbitration International, 28, 2012, 591, 637-49.

114
This was the role that the German BGH took. Strong, Class, Mass, and Collective Arbitration (n 1) 18-19, 86. Courts in the United States have been less forthcoming about the procedural minimums that must be met, although that is likely because class arbitration closely resembled judicial class actions, which had already been determined to be procedurally fair. ibid 38-40, 43.

115
This was the approach used in Spain and, to a lesser extent, the United States. In Spain, the legislature created a specific procedural statute outlining every nuance of the procedure. In the United States, the legislature approved the rules of court (ie the Federal Rules of Civil Procedure) that outline judicial class action procedures. ibid 276. Although the legislature did not explicitly extend those rules to arbitration, legislative intent was to some extent reflected in US class arbitration procedure, since common law and rules-based class arbitration essentially mirror US class action procedure. ibid 38-40, 43, 276.

116
ibid 12.

117
ibid 240, 304, 332-33.

118
ibid.

119
ibid 331 (discussing the resolution of certain Holocaust-related claims).

120
For example, the tribunal in Abaclat considered adopting some of the evidentiary procedures used by mass claims commissions. ibid 332.

121
(n 10) (discussing mass and multiparty claims involving Argentina and Cyprus).

122
US 444 (2003).