Advantages and Disadvantages of Bellwether Cases

Bellwether Trials – Introduction

At a time when employment of class actions within the legal sphere is increasingly being compounded by skepticism, the current multidistrict litigation (MDL) process that is defined in 28 U.S.C 1407 is gaining prominence as the most viable way of resolving complex civil cases. This was created in 1968 by the Congress and its main aim was to provide a central forum for handling civil cases. In particular, it provided a platform upon which various common questions were consolidated in order to enhance the effectiveness and efficiency of pretrial motions as well as discovery proceedings[1]. From a theoretical point of view, the central forums, also known as transferee courts, provide a viable environment where preliminary aspects pertaining to any litigation are addressed prior to sending individual cases to various destinations for resolution. Practically, the transferee courts perform various other purposes in addition to acting as a discovery point. In this regard, the court develops viable ways for conducting the representative or bellwether trials and thereby enhances and speeds up the MDL process as well as global resolutions that transpire along the way.

The purpose of this paper is to explore the advantages and disadvantages of bellwether cases and to review how the cases are chosen to ensure fair representation. To enhance a harmonic consideration, it begins by detailing the historical background of bellwether cases. In particular, it reviews early attempts that were geared towards binding related claimants to trial verdicts and the modern non binding approach that simply benefits from the information that the cases generate. Basically, a bellwether cases in most cases begin as a normal lawsuit. It then proceeds to the pretrial discovery stage and later to trial using a single plaintiff as well as defendant.  The case qualifies to be a bellwether when its defenses, claims and facts are similar to those of the wider group of other related cases. The information approach in this regard is based on the recognition that the outcomes of the bellwether cases do not necessarily need to be binding of related cases for them to benefit the MDL process.

By incorporating fact finding and jury influence in the MDL, the bellwether cases are instrumental in enhancing maturation of disputes. This is so because they provide the coordinating counsel with the opportunity to coordinate the pretrial common discovery products analyze the strengths as well as shortcomings of their evidence and arguments and understand and appreciate the costs as well as risks that are related to the litigation. Typically, the bellwether process needs to culminate in the formation of ‘trial packages’. These are employed by the local counsel after the MDLs have been dissolved. Of great importance however is the experience and knowledge that is gained during the bellwether trial process.  This has the ability to enhance global settlement negotiations. In addition, they ensure that the respective settlement process does not take place in a vacuum but in an actual environment characterized by analysis of the respective litigation process by a host of juries.

Then, the paper proceeds to evaluate the various practical considerations that the counsel and the courts employ during bellwether trials. In particular, it underscores the methods that the preceding parties employ in selecting representative bellwether cases from the universe. The trial processes are wide and varied although an ideal method needs to ensure that the cases are representative of the entire claimants. This paper analyzes the various methods in a bid to identify the most viable method that would ensure that bellwether cases truly perform their roles as such.

Development of Bellwether Trials

For a significant period of time, the federal courts and the state have persistently grappled with the unusual problems that stem from mass tort litigation[2]. To address the issues in an effective manner, the courts are currently opting to use simple as well as complex creative procedures like the bellwether trial[3].  The concept ‘bellwether’ is derived from the tradition of using a rope to fasten a bell around the castrated male sheep’s neck that is usually chosen to act as a lead or head of the rest of the flock[4]. Essentially, a bellwether is a pacesetter of all the trends or key determinant of the practices[5]. In the legal context, a trial is considered a bellwether when the verdicts of constituent sample cases are employed for resolving other cases[6].  Currently, bellwether trials are informally employed by the judges in mass litigation[7]. In this regard, they are useful for valuing the cases as well as encouraging settlement.  Regardless of the fact that the origin of the concept of ‘bellwether’ is compounded by certain complexities, employment of bellwether trials especially in mass tort litigation has increasingly gained momentum.

Fundamentally, the courts seek to use the outcomes of bellwether trials to formally bind related claimants[8]. In a binding bellwether trial procedure, the courts begin by choosing certain cases that they seek to present to the jury during trial. The judge then divides the selected cases in two phases; liability and damages. Alternatively, the judge might decide to classify these in three categories of causation, liability or damages. After this, the court proceeds to trying the bellwether cases in stages while the jury provides a verdict of each case at each stage. In the end, the outcomes of the trials are inferred to the rest of the plaintiffs.[9]  At the outset, courts attempted to use binding bellwether trials as type of class action adjudication. [10]  Despite the fact that there was no class certification, courts employed the binding approach on the premise that trying representative cases was sufficient to have a binding impact on all the related cases.  However, the appellant courts doubted the credibility of the binding approach.  For instance, in Cimino v. Raymark, the United States Court of Appeals for the Fifth Circuit refers to the majority opinion in earlier Fifth Circuit case, In re Chevron U.S.A., Inc., to clarify that Judge _____’s language favoring the binding use of statistically representative bellwether plaintiffs is “plainly dicta, certainly in so far as it might suggest that representative bellwether verdicts could properly be used to determine individual causation and damages for other plaintiffs.”[11]  Other circuits have also recognized that the results of bellwether trials are not properly binding on related claimants unless those claimants expressly agree to be bound by the bellwether proceedings. For instance, in In re Hanford Nuclear Reservation Litigation, the Ninth Circuit held that the results of the Hanford bellwether trial would not be binding on the remaining plaintiffs.[12]  Along the same lines, in In re TMI Litigation, the Third Circuit held that “absent a positive manifestation of agreement by Non-Trial Plaintiffs, we cannot conclude that their Seventh Amendment right not compromised by extending a summary judgment against the Trial Plaintiffs to the nonparticipating, non-trial plaintiff.”[13]  Such experimentation of the binding strategy by courts did no last for long. It was terminated by the decision by the Fifth Circuit which deemed the practice unconstitutional based on the provisions of the Seventh Amendment.[14]

Today courts use no longer use bellwether trials for the purpose of resolving thousands of related cases pending in a MDL in one “representative” proceeding, but rather to provide meaningful information and experience to everyone involved in the litigations – the nonbinding informational approach.  Although cases addressing a particular issue are chosen and reviewed, their outcomes or verdicts are non-binding and do not influence the rest of the litigants in the respective category in any way.  Usually, relative outcomes are useful in helping the parties to arrive at a common and undisputable settlement. Alternatively, the respective parties are free to ignore the outcomes completely and instead undertake individual trials.

Advantages of Bellwether Trials

Bellwether trials benefit the litigation process as well as the involved parties in different ways.  As aforementioned, bellwether trials within the MDL context can be effectively employed for nonbinding informational reasons. For instance, they can be useful for testing a host of theories, strategies and defenses in a real live trial. In this respect, bellwether trials are useful for enhancing the experience, capacities and competencies of the involved parties.[15]  While the outcomes of such bellwether trials can only bind the specific bellwether cases, they can still benefit the related claimants and the MDL procedure. The Fifth Circuit best illustrates the importance of bellwether trials in this regard:

The notion that the trial of some members of a large group of claimants may provide a basis for enhancing prospects of settlement or for resolving common issues or claims is a sound one that has achieved general acceptance by both bench and bar …. The reasons for acceptance by both bench and bar are apparent.  If a representative group of claimants are tried to verdict, the results of such trials can be beneficial for litigants who desire to settle such claims by providing information on the value of the cases as reflected by the jury verdicts[16].

Also worth recognizing is that bellwether trials provide a vehicle upon which litigation theories can be effectively put in practice.  However, it is well known that the trial environment is not only dynamic but very complex and as such, trials generally do not proceed as intended by the parties or their respective attorneys. Besides the unforeseen logistical issues, the effects of evidence and facts upon the court and jury remain uncertain. During multidistrict litigation, the preceding uncertainties are usually made worse by the differences that exist among the circumstances of consolidated claimants and by the volume of pertinent evidence produced during the course of the discovery.  Bellwether trials aid in alleviating such uncertainty and by giving the coordinating counsel the chance to organize the products of pretrial discovery, assess the strengths and weaknesses of their arguments, strategies, and evidence, and internalize the risks and costs involved in the trial process.[17]  In the long run, all parties are not only prepared but also well informed about the litigation process. Certainly, the employment of bellwether trials is beneficial and has the capacity to speed up and improve the MDL process in two main ways.    Firstly, they give the coordinating counsel a chance to perfect their presentations for future trials, potentially resulting in the creation of “trial packages” that will likely prove useful to local counsel upon the dissolution of MDLs.   Secondly, by indicating future trends, such as how a certain claims may fare before the jury, bellwether trials have the potential to inform and trigger settlement negotiations.[18]

The Trial Packages

Bellwether processes benefit the MDL process by motivating the attorneys to develop ideal trial packages. As indicated before, bellwether trials compel the litigants to streamline, restructure and systemize comprehensive material and information that are generated during the pretrial discovery phase of the MDL. These packages are an invaluable resources and can be employed by local counsel or litigants once the MDL is dissolved and hen the individual cases have been remanded for trial to transferor courts. A series of bellwether trials give the counsel an opportunity to perfect their presentations by making minor adjustments usually based on litigation realities or on previous performances.

In the long run, trial package safeguard the knowledge and information that is acquired or developed by the coordinating counsel. This is particularly important because the respective knowledge is likely to be lost in instances when global resolution my no be attained by the transferee court. In instances where the cases under trial are remanded, trial packages are useful as they ensure pretrial common discovery products do not override the capacity of the coordinating counsel. Certainly, the bellwether processes ensures the effective functioning of the transferee court and eventual attainment of its goals with regards to preparing the cases for trial at the local district level efficiently and simplifying pretrial discovery.

Global Settlement

Essentially, the goal of a mass tort trial is to attain a settlement through harmony. In light of the MDL Panel, the transferee court provides the most viable environment for global settlement negotiations. This is because it provides room for a central forum comprising of lawyers and litigants from across the nation. These assemble before one judge and the settlement process takes place ago. Transferee courts can play a leading role in necessitating fulfillment of relative goals by creating and managing bellwether trials. Nonetheless, it should be appreciated that the process can be undermined counter arguments by the appellate[19]. In this respect, some courts consider bellwether trials to be akin to judicial blackmail[20].

In sum, the concept of bellwether trial process stemmed from the custom of tying a bell around the male sheep’s neck. This male sheep was considered to be the leader of the flock and the bell was used as an identification of the respective flock. Likewise, the verdicts of bellwether trials were employed as determinants of the fate of a host of claimants. During their experimentation, the credibility bellwether trials were doubted by the appellate and the Fifth circuit demanded their prohibition due to the fact that they were unconstitutional. Regardless of this, the bellwether trials have various benefits to the to the MDL process. As indicated in the preceding review, they are not only employed as non binding informational approaches but they also aid in testing of theories. Most importantly, they help in creation of the trial packages that ease the litigation process in various ways. Nevertheless, it should be acknowledged that they still face opposition from the appellate that considers them to be forms of judicial blackmail.

• The Bellwether Selection Process

After the threshold determination to utilize bellwether trials, the transferee court and coordinating counsel should focus on the mechanics of the trial selection process.  The process of selecting representative bellwether plaintiffs is a highly controversial issue in mass tort proceedings.  Bellwether trials must produce valuable information that will allow the parties involved to evaluate the strength and settlement value of all the related cases.   In order for bellwether trials to fulfill their valuable purpose of being “informative indicators of future trends” and “catalysts for an ultimate resolution”, it is critical that the transferee court and attorneys involved formulate a trial selection process that will be fair to all parties involved and most importantly render a representative set of bellwether plaintiffs.[21]  Thus, selecting cases for bellwether trials and constructing litigation strategies require thoughtful consideration by both plaintiffs and defendants.  However, for this to occur it is imperative that there be a sufficient number of cases tried, and that the cases selected be representative of the range of cases, in order to enable the parties to determine what range of value the cases may have if resolved in the aggregate. [22]  For it is only when a “representative … range of cases” is selected may “individual trials … produce reliable information about other mass tort cases.”[23]  The Manual for Complex Litigation instructs that in the selection of bellwether cases

Sampling and surveying can be used to obtain information useful both for settlement and for bellwether trials of the sample cases or for a class trial.  Whether the aim is settlement or trial, the court should ensure that the sample is representative of all claims encompassed in the particular proceedings with respect to relevant factors, such as the severity of the injuries, the circumstances of exposure to the product or accident, applicable state law, and the products and defendants alleged to be responsible.[24]

Ideally, the trial selection process should accurately reflect the individual categories of cases that comprise the MDL, illustrate the likelihood of success and measure of damages within each respective category, and illuminate the forensic and practical challenges of presenting certain types of cases to a jury.  Any trial selection process that strays from this path will likely resolve only a few independent cases and have a limited global impact.[25]

Before selecting the actual bellwether, the transferee court and coordinating counsel must take preparatory steps to ensure the selection of a representative sample of bellwethers.  First, the transferee court and attorneys must catalogue the entire universe of cases that comprise the MDL and then divide the cases into several distinct, easily ascertainable categories of cases, meaning they must ascertain the makeup of the MDL. The rationale behind cataloguing and dividing the entire universe of cases within the MDL is simple. A bellwether trial is most effective when it can accurately inform future trends and effectuate an ultimate culmination to the litigation; therefore, it is imperative to know what types of cases comprise the MDL.  Otherwise the transferee court and the attorneys risk trying an anomalous case, thereby wasting substantial amounts of both time and money.  Thus to ensure that the cases ultimately tried are emblematic of all the cases comprising the MDL, the transferee court and the attorneys must determine the composition of the MDL prior to engaging in any further trial selection steps.  To discharge this task effectively, the transferee court and the attorneys should each conduct a census of the entire litigation and identify all the major variables.  This initial step in the bellwether process will require that the attorneys have some knowledge about the individual cases in the MDL.  In the Vioxx MDL, this was achieved with limited case-specific discovery through the exchange of plaintiff and defendant profile forms.[26]

In any given MDL, there will be innumerable variables differentiating each case from the others.  Rather than attempt to delineate every identifiable variable, the transferee court and the attorneys should focus on those variables that can be easily identified, are substantively important, and provide clear lines of demarcation – i.e., the major variables.  To put it summarily, these groupings will act as guideposts, focusing the attorneys on the most predominant and important issues in the litigation.  After the transferee court and the attorneys have each separately evaluated the composition of the MDL and considered all the major variables, the transferee court should hold a status conference at which time it and the attorneys should discuss all of the relevant variables in an attempt to reach a consensus on which variables are the most predominant and important.  By the conclusion of this status conference, the court should determine how the MDL will be divided and, more importantly, the attorneys should know why the groupings have been chosen.

After determining the composition of the MDL and creating groupings by which to divide the MDL, the transferee court and coordinating counsel should begin the process of creating a manageable pool of cases from which the bellwether cases will eventually be selected.  This pool of cases will reflect the various categories and contain cases that are both amenable to trial in the MDL and close to being trial-ready.  Once the trial selection pool has been assembled, each of the cases within the pool must undergo case-specific discovery.  This discovery process will typically be no different from that which occurs in an ordinary case, and thus requires no additional explanation here.  Near the conclusion of the case-specific discovery, the transferee court and attorneys will select the bellwether cases.   The next section will explore the various bellwether selection methods that have been used, in an effort to determine which method is likely to produce the most representative set of bellwether plaintiffs.

A. Selecting the Bellwether Plaintiffs

There are essentially four basic approaches to selecting bellwether plaintiffs: (1) random selection, (2) selection by transferee judge, (3) selection by plaintiffs’ counsel, and (4) selection by plaintiffs and defense counsel. [27]  Additionally, the transferee court may allow the attorneys to exercise a predetermined number of strikes or vetoes to eliminate potential bellwether cases in the pool, from consideration prior to the actual selection.[28]   While all these selection methods have been utilized over the years, mere usage does not prove/confirm their reliability.  Rather, at issue is whether any of the selection methods reliably produces a true bellwether verdict.  A bellwether verdict generated by an improper selection method has the potential to skew not only the verdict but also any results influenced by it.  It is always important to keep in mind the goal – the selection of typical cases “which when decided and reviewed may provide a legal and factual pattern against which the remaining issues in the pending cases may be subsequently matched.”[29]  In essence, bellwether verdicts should be used to “develop a pattern for evaluation.”[30]   Each of the four bellwether selection methods will be evaluated below.

1. Random Selection

Legally, it is widely agreed that for bellwether cases to be valid and representative of the population, they need to be not only sufficient in number but also selected randomly[31]. The Manual for Complex Litigation accredits random selection to be the most ideal method of attaining effective representation especially during identification of cases. According to it, the credibility of bellwether cases is greatly determined by equal representation of the plaintiffs and well as their claims in the cases employed as samples. Further, it cites that the tendency of some judges to allow the plaintiffs and defendants to determine the cases needed to undergo initial trial undermines the credibility of the end result. The manual insists that effective or equal representation of cases can only be attained if the judge directs the relevant parties to employ random selection or use a couple of cases that are considered by all parties to be an ideal mix[32].

Random selection is a valid and credible strategy that is usually pursued in two main ways. It can be undertaken in a direct manner that is commonly referred to as ‘out of hat’[33]. Alternatively, more sophisticated or complex methods can be employed in choosing the sample[34]. Regardless of the type of methods employed, it is generally agreed that random sampling is not only a fair approach but it is also rational because of the fact that each case in the population is given an equal chance to be a part of the sample[35]. In addition, it is simple, devoid of in depth analyses by the attorneys or judges, straight forward and makes sure that ideal cases that are representative in nature are chosen[36]. Also worth mentioning is the fact that random sampling helps to filter the cases by separating the credible cases from fraudulent ones and thereby reducing relative costs. In this regard, costs are saved when the fraudulent cases are exposed to intense scrutiny and investigation to ensure that selected cases are based on credible grounds[37]. Nonetheless, random selection is also compounded by two main shortcomings.

To begin with, random selection has the potential to alienate attorneys from the entire procedure. In this respect, it is worth appreciating that defeats and victories are not the sole reasons why bellwether cases are tried. The process of trial gives the attorneys a chance to prepare their cases prior to trial and familiarize themselves with wide ranging vital decisions that need to be made upfront. Allowing attorneys to participate in choosing the cases that they would try gives them an opportunity to further their individual selfish interests. In essence, random selection prevents coordinating attorneys from exploring various methods of trial and determining their practicality before a jury, hence inhibiting a possibility for mass resolution at the end of the trial[38]. Secondly, it is posited that random selection culminates in a possibility of unequal representation of bellwether plaintiffs in the sample.  Ideally, a typical random sample is likely to comprise of two of the five needed variables. Undoubtedly, this undermines the credibility of the entire process.

However, the risk of having an unrepresentative set of bellwether plaintiffs may be remedied by using scientific statistical evidence, also known as stratified random sampling.  Some circuits have recognized the use of inferential statistics with random sampling as a valid method of selecting bellwether plaintiffs.[39]  In Cimino v. Raymark Industries, Inc., Judge Parker used statistical sampling by Professor Ronald G. Frankiewicz to select the representative plaintiffs whose claims would be tried in the asbestos litigation.[40]  Dr. Frankiewicz attested to the use of stratified random sampling, an approach used by the United States Census Bureau that sets the total group of plaintiffs as component populations, each of which is sampled using a simple random plan.[41]  He suggested that the plaintiffs be stratified into smaller groups according to (i) primary medical complaint and (ii) severity of the medical condition, and sampled at random to attain a group of representative bellwether plaintiffs.[42]  The use of statistical random sampling enabled Judge Parker to establish that his plaintiff selections were reliable to a 99% confidence interval.[43]  Furthermore, such statistical random sampling has been approved by significant legal scholars such as Professor Saks and Blanck, who describe the procedure as follows:

Mass torts represent a sampling theorist’s dream.  The population of cases from which the sample is to be drawn is known with unusual completeness.  This provides the sampling frame from which any type of case sampling proceeds.  In addition, many details are known or can be learned about each member of the population.  Thus, the degree to which the sample is representative of the population can be known with near certainty – a great improvement over most sampling situations.  Representativeness is the touchstone of good sampling.

[Concluding sentence]

Test cases should produce a sufficient number of representative verdicts to enable the parties to determine what range of value the cases may have if resolved in the aggregate.  Manual for Complex Litigation (4^th) § 22.315.  If the cases are randomly selected, they will reflect the full range and diversity of the claims at issue, including in terms of factual issues, choice of law, legal theories and defenses, and perhaps counsel as well.

2. Selection by the Transferee court

This is a step wise process that begins by the attorneys either individually or jointly preparing reports about every case included in the trial-selection pool. These reports outline various factors ranging from the facts that are related to each case (both those that are contentious and those that have been agreed on) to the main legal issues surrounding each case and their personal positions regarding why each case needs to be selected or exempted from being employed as a bellwether case by the respective transferee court. The main advantage of this approach is that it allows the transferee court to make sure that all predetermined variables are analyzed by all parties at the trial. This ensures that the ultimate selection is fair and just to both sides.

The main weakness of this approach pertains to its tendency to minimize attorney participation. However, compared to random selection, this allows the attorneys to present an argument for or against the selection of the respective cases. It gives them a chance to voice their individual reasons pertaining to why they think certain cases need to be tried or not tried. However, it should be appreciated that allowing the attorneys to voice their personal views regarding the cases does not mean that the respective views would be effected. Also, in some instances, attorneys may not be willing to voice their internal motives or share their personal views with the opposing counsel or transferee court. Moreover, the process is likely to be time consuming and hence may not be appropriate for attorneys and the transferee court. This is because it requires the affected attorneys to prepare comprehensive reports for every case. The transferee court on the other hand is required to critically evaluate the merits or advantages of each case.

Being an unbiased neutral, the transferee court’s selections are likely to be more focused on cases that are truly representative of the litigation and not on cases that present the best opportunity for success at trial.  For instance in In re Baycol Prods. Liability Litigation the court determined that it will select the bellwether cases if the parties are unable to agree.[44]

Although the existence of a neutral arbiter is undoubtedly a great benefit, it is highly unlikely that the transferee court can properly conduct this task on its own.  Given their inherent costs, bellwether trials will generally only be utilized in large-scale MDLs.  Such MDLs typically consist of thousands of individual cases.  Some cases will be filed directly in the transferee court.  Some will be filed in, or moved to, other federal district courts and then transferred to the transferee court by the MDL Panel.  Still others may be pending in state court awaiting trial.  The transferee court simply does not have the resources available, or the familiarity with each individual case, to conduct this task adequately.  Even if the attorneys prepare briefs outlining the potential cases, similar to a final pretrial order, it is still doubtful that the transferee court’s selections will be as knowledgeable as the attorney’s picks.  Therefore this option should also be avoided.

3. Selections by Plaintiffs’ Counsel

Another selection strategy is to allow the plaintiffs’ counsel to select the bellwether cases.  This reasoning is based on the premise that if one side only is allowed to select all the required bellwether cases and the respective side eventually looses all or a significant percentage of the selected trials, it can be concluded that the theories employed by the respective side are not credible.  This increases the likelihood that the litigation will be resolved quickly and easily.  This strategy was employed in Propulsid MDL.[45]   Essentially, this approach is efficient because it employs only a single side during the bellwether trial selection and it provides the plaintiffs’’ counsel the capacity to actively participate in the selection process.

While it would seem self-evident that a trial of plaintiffs selected by one side only will not provide meaningful information (unless the other side manages to prevail anyway), it is not unusual for courts to allow plaintiffs to select the “test” cases, as has occurred in aspects of the Vioxx mass tort.[46]    Additionally, the Welding Rod MDL addressed the defendants’ concerns that the bellwether plaintiffs selected for trials had been almost entirely of the plaintiffs’ choosing.[47]  There the defendants urged that the bellwether plaintiffs should have been selected randomly by the court.  While the court recognized the defendants’ concerns, it held that the bellwether selected by plaintiffs were sufficiently representative to be valid choices for bellwethers.   The court noted that the purpose of the bellwether trials was to produce a sufficient number of representative verdicts to enable the parties and the court to facilitate a resolution for the entire body of cases.

The disadvantage of this approach is that it greatly favors the plaintiffs’ side at the expense of the defense.  The credibility of the process may be undermined by the fact that the plaintiffs’ side is likely to select the cases that further their ends simply because they have the ability to do so.  In the process, they may omit important variables from a sample that would be representative of the entire population.   This opens the door for the inequitable stacking of overtly unfavorable and possibly unrepresentative cases, as well as creating an atmosphere of antagonism.

Moreover, there is also the risk that a few early bellwether trial results could have significant unfortunate effects of decreasing rather than increasing the prospects of any early global resolution, unreasonably raising the expectations of the plaintiffs’ bar.  Under this approach, there is the serious concern that the plaintiffs will secure verdicts in their selection of a few best cases out of the hundreds or thousands which bear no resemblance to the best few bellwethers.  This may not encourage rational defendants to alter their views about the merits of the bulk of cases.

4. Selection by Plaintiffs and Defense Counsel

The final approach is involving both sides of the coordinating in the selection of the bellwether cases.  There are two ways to accomplish this.  First, the transferee court can require that the plaintiffs and defense counsel to jointly agree on which bellwether plaintiffs to select.  The rationale is that if the sides can agree on the cases, the cases will likely be representative and fair to both sides.  Ideally, it would probably be best if the attorneys responsible for coordinating the process mutually agree on the cases that need to be included in the bellwether trials.  Irrespective of the fact that this method may be the most desirable, it is not achievable from the practical point of view.  This is because the relative stakes are likely to be enormous and their values so divergent that it becomes difficult for the attorneys to arrive at a mutual agreement regarding the cases.  Further, by being stubborn in their advocacy, as opposed to participating in meaningful, good-faith negotiations, the attorneys will lose an opportunity to resolve their clients’ cases effectively and efficiently.  An example that best illustrates this is the Vioxx Litigation where the first case only was agreed upon and ultimately selected.[48]

The second option requires that both coordinating attorneys participate in the selection process by alternating picks.  By giving both sides an equal chance to participate in the selection process, this approach enhances fairness.  Further, it offers the benefit of allowing each side to test out issues that are important to its strategy, or which each party feels will be persuasive to the other side once resolved.  Most importantly, it is efficient and ensures that professional and competent personnel assume the noble responsibility of selecting trial cases.  In the Vioxx MDL case, each side of coordinating attorneys was allowed by the transferee court to make five choices.[49]  Then, each side was permitted to veto two cases from the list of its opponents.[50] Finally, the six cases that remained were tried on a rotation basis, beginning with the selection of the plaintiff.[51]  The fact that both attorneys are involved however undermines the efficiency of the process.  However, the primary problem with this method is that by allowing each side to select half of the bellwether plaintiffs, the sample may not be representative, but rather a mix of the best and worst cases for each side.[52]  Mainly it does not eliminate or minimize the chance that the attorneys will select favorable, rather than representative cases.

Conclusion

Plagued by protracted trials and in an effort to better serve judicial economy, courts are increasingly experimenting with utilizing representative plaintiffs for determining a defendant’s liability to the class as a whole in mass, complex, toxic tort cases.  All of this experimentation occurs amid questions about the effectiveness of the representative plaintiffs’ approach in accomplishing the goal of setting benchmarks for settlement.  However, as this debate continues, courts increasingly are required to establish a criterion for selecting such representatives.[53]

At this point, it is certain that this approach offers the most credible method of selecting bellwether cases. Thus to ensure representativeness of the bellwether sample, both sides of the coordinating counsel need to evaluate the strengths and shortcomings of the selected cases.

• [1]  28 U.S.C. § 1407 (2000)
• [2]  See Manual for Complex Litigation § 38.2 (3d ed. 1995)
• [3]  See id.
• [4]  In re Chevron U.S.A., Inc., 109 F.3d 1016, 1019 (5^th Cir. 1997)
• [5]  Bellwether Trials by Lahav
• [6]  Bellwether Trials by Lahav
• [7]  See, e.g., Cimino v. Raymark., Indus., Inc., 151 F.3d 297, 318 (5^th Cir. 1998).
• [8]  Bellwether Trials by Lahav, p6
• [9]  See, e.g., Cimino v. Raymark., Indus., Inc., 151 F.3d 297, 318 (5^th Cir. 1998).
• [10]  See id.
• [11]  Cimino v. Raymark Indus., 151 F.4d 297 (5^th Cir. 1998)
• [12]  In re Hanford Nuclear Reservation Litig., 497 F.3d 1005, 1025 (9^th Cir. 2007)
• [13]  In re TMI, Litig., 193 F.3d 613, 725 (3d Cir. 1999)
• [14]  Cimino v. Raymark.
• [15]  Bellwether Trials in Multidistrict Litigation, p 2337
• [16]  In re Chevron U.S.A., Inc., 109 F.3d 1016, 1019 (5^th Cir. 1997)
• [17]  Bellwether Trials in Multidistrict Litigation, 2338
• [18]  Bellwether Trials in Multidistrict Litigation, 2338
• [19]  Castano v. American Tobacco Co., 84 F.3d 734, 746 (5^th Cir. 1996)
• [20]  Bellwether Trials, p2343
• [21]  Bellwether Trials in Multidistrict Litigation, p2343
• [22]  Manual for Complex Litigation (Fourth) §22.315(2004)
• [23]  Manual for Complex Litigation (Fourth) §22.315 (2004)
• [24]  Manual for Complex Litigation (Fourth) §33.27 at 326.
• [25]  Bellwether Trials, p2343
• [26]  See In re Vioxx Prods. Liab. Litig., 501 F. Supp. 2d 789, 790-91 (E.D. La. 2007).
• [27]  Sterling v. Velsicol Chem. Corp., 855 F.2d 1188, 1196 n.6 (6^th Cir. 1988) (declining to address the issue because it was not raised as error on appeal) (emphasis added).
• [28]  Bellwether Trials in Multidistrict Litigation.
• [29]  Allen v. United States, 588 F. Supp. 247, 258 (D. Utah 1984), rev’d on other grounds, 816 F.2d 1417 (10^th Cir. 1987)
• [30]  Manual for Complex Litigation, §23.15, at 173 n.486.
• [31]   James M. Wood, the Judicial Coordination of Drug and Device Litigation: A review and Critique, 54 FOOD DRUG L.J. 325, 347 (1999).
• [32]  Manual for Complex Litigation (Fourth) § 22.315 (2004)
• [33]  See In re Prempro Prods. Liab. Litig., MDL No. 1507 (E.D. Ark. July 14, 2005) (order regarding bellwether trial selection); In re Prempro Prods. Liab. Litig., MDL No. 1507 (E.D. Ark. July 14, 2005) (letter order).  Information about these two orders can be found on a Web site dedicated to this multidistrict litigation.  See Prempro Product Liability, https://www.are.uscourts.gov/mdl/index.cfm (last visited June 13, 2010)
• [34]  In the Bextra & Celebrex MDL, the transferee court had attorneys use a third-party randomizer computer program as a random selection method.  See In re Bextra & Celebrex Mktg. Sales Practice & Prod. Liab. Litig., MDL No. 1699 (N.D. Cal. Nov. 17, 2006) (pretrial order no. 18) (describing the initial selection of plaintiffs for discovery and trial pool), available at https://ecf.cand.uscourts.gov/cand/bextra/content/files/pretrial_order_18.pdf
• [35]  See Manual for Complex Litigation (Fourth), § 22.315
• [36]  See, e.g., In re Chevron, 109 F.3d at 1019 (“A bellwether trial designed to achieve its value ascertainment function … has a core element representativeness – that is, the sample must be a randomly selected one.”)
• [37]  From Both Sides Now: Additional Perspectives on “Uncovering Discovery”.  By Amy Schulman and Sheila Birnbaum
• [38]  Moreover, although random selection may be theoretically attractive method for selecting bellwether trials, and, indeed, although some courts and commentators have suggested that it may even be of constitutional significance when the results of bellwether trials are used to bind related claimants, random selection is of considerably less importance when bellwether trials are employed in practice for nonbinding informational purposes.
• [39]  See In re Chevron USA, Inc., 109 F.3d 1016, 1017-18 (5th Cir. 1997); In re Estate of Marcos Human Rights Litigation, 910 F.Supp. 1460, 1467 (D.Haw. 1995)
• [40]  751 F. Supp. 649, 664 (E.D. Tex. 1990).  Judge Parker’s opinion in Cimino contains an extensive and scholarly discussion of the history and reliability of these statistical procedures.  See id. at 659-66.  Essentially, Professor Frankiewicz’s procedures are acceptable for the “same reasons that society embraces the science [of statistics].  It has been proved to provide information with an acceptable degree of accuracy and economy.” Id. at 663; see also In re Estate of Marcos Human Rights Litig., 910 F. Supp. 1460, 1464-68 (D. Haw. 1995) (using stratified random sampling to select representative trial plaintiffs).
• [41]  See Affidavit of Ronald G. Frankiwicz, Ph.D. at 4, Adams (No. H-96-1462) (on file in Appendix, supra note 17, at 20)
• [42]  See id. at 7-9.
• [43]  See Cimino, 751 F. Supp. At 664.
• [44]  In re Baycol Prds. Liab. Litig., MDL No. 1431 (D. Minn. July 18, 2003) (pretrial order no. 89 at 2) (providing that the court will determine eligible cases to be tried if the parties are unable to agree), available https://www.mnd.uscourts.gov.Mdl-Baycol/pretrial_minutes/baycol89.ord.pdf.
• [45]  See In re Propulsid Prods. Liab. Litig., MDL No. 1355, 2003 WL 22023398, at *1 (E.D. La. Mar. 11, 2003)
• [46]  In re Vioxx Lit., Case No. 619 (N.J. Super. Ct., Atlantic City) (ordering plaintiff to select a grouping of eight plaintiffs, with trials consisting of two or more plaintiffs)
• [47]  In re Welding Fume Products Liability Litigation 2007 WL 1702953 (N.D. Ohio June 26, 2007)
• [48]  See discussion supra note 107.
• [49]  See In re Vioxx Prods. Liab. Litig., 501 F. Supp. 2d 789, 791 (E.D. La. 2007).
• [50]  See id.
• [51]   See id.
• [52]  In re Chevron USA, Inc., 109 F.3d 1016, 1019 (5^th Cir. 1997)
• [53]  Sterling v. Velsicol Chem. Corp., 855 F.2d 1188, 1196 n.6 (6^th Cir. 1988) (declining to address the issue because it was not raised as error on appeal) (emphasis added).