We wrote recently on Lone Pine orders and the role they might play in today’s world of inventory litigation. You know what we are referring to – the mass litigation that occupies much of our time, where some people would say that the number of cases involving similar products is more important than whether any particular case has arguable merit. We are not among the people who subscribe to that view, but it got us to thinking about resolution, and particularly about trials, and particularly about bellwether trials, which have become common in multidistrict litigation and other coordinated pretrial proceedings. Bellwether trials are supposed to result in bellwether verdicts, which are intended to provide benchmarks against which to measure the resolution value of similar cases.
It is an imperfect concept, which we have noted multiple times (including here and here ). But it is the system that we have, and it has gained both legitimacy and traction. Following a number of trials in the Vioxx MDL – an MDL that saw the resolution of a great many cases – Judge Fallon and his clerks published an article in the Tulane Law Review (available here) providing very thoughtful commentary on the entire process. For our part, we have participated in the selection and trial of bellwether cases, and sometimes multiple claims get resolved in the aftermath, and sometimes they do not.
Which leads to our question for today: Has the bellwether trial process lost its luster? We have not yet formed a strong opinion one way or another, but we raise the issue now because of an order entered earlier this month in the Cook Medical pelvic repair system MDL in the Southern District of West Virginia. See Pretrial Order #59, In re Cook Medical, Inc. Pelvic Repair System Prods. Liab. Litig., MDL No. 2440, 2015 WL 3385719 (S.D. W. Va. May 19, 2015). There, the district court set up a bellwether trial process that largely followed the approach that Judge Fallon and his co-authors set forth in their article. The court first selected a pool of 30 cases to be worked up in discovery, with each side selecting 15 cases. That was about 20 percent of the then-pending cases, which is a significant number, and from that group the district court selected four “bellwether cases” that would be tried to verdict and provide the benchmarks that the bellwether process aims to provide. (Id. at pp. 1-2)
The proceedings, however, did not go as the court expected. Before trial, the plaintiffs voluntarily dismissed all four bellwether cases with prejudice. Returning to the bucolic metaphor from which the “bellwether” tag originates, that left no neutered ram wearing a bell to lead the flock. To make matters more frustrating for this district court, the plaintiffs dismissed 20 other cases in the original discovery pool, leaving just six cases. And in four of those remaining six cases, plaintiffs’ counsel moved to withdraw rather than oppose the defendants’ motions for summary judgment based on lack of causation. (Id.)
Thus, from the initial pool of 30 cases, due to the plaintiffs’ machinations, the court was left with just two, leading the court to conclude that “I am once again finding that the bellwether process is not effective. . . . While I had hoped that representative cases had been chosen in the bellwether process, without a trial of those cases because of their dismissal with prejudice by plaintiffs before trial, the parties have no opportunity to confirm that they were representative or determine potential values of these cases and the true cost of working them up for trial.” (Id.. at p. 2) The court’s solution was to order upward of 250 cases into discovery in preparation for remanding the cases to the transferor districts for trial. (Id.) That is a strident reaction, and we can’t help but wonder whether the court pulled the plug on the bellwether process too soon.
This is a thought-provoking order for a couple of reasons. First, whether the bellwether selection process was succeeding in this particular instance depends on how you define success. Sure, the process did not result in any verdicts. But merely by putting 30 cases into the process and shining the light on their case-specific facts, the court got rid of all but two of them. If resolution is the goal, that is a 93 percent “success” rate. Remember our reference to Lone Pine at the outset? Maybe if such an order were in place at the beginning for all the cases, the culling effect would have been comparable. We do not know why 24 plaintiffs dismissed their cases, but we suspect that the 28 dismissed/abandoned cases should not have been filed in the first place. Remember also that 15 of the cases were plaintiffs’ picks.
Second, the court had hoped that the bellwether trials would have confirmed the “representative” nature of the selected cases. We do not overestimate the ability of a bellwether selection process to select “representative” cases, and we doubt that the district court did either. The fact is that even when dealing with similar products, no two cases are alike, and resolution of one does not necessarily signal the value of another. This is the core reason why personal injury class actions no longer exist, and if what we have in their place is inventory litigation led by bellwether trials, the best we can expect are approximations. Helpful approximations in many cases, but approximations nonetheless. Here, the approximation is that 93 percent of the Cook cases have zero value, in the estimation of plaintiffs’ counsel themselves. That should tell everyone something.
We again have no strong opinion on how this particular MDL should go forth, but the district court’s order touches on two cracks in the bellwether process. Plaintiffs hold a card up their sleeves—the ability to dismiss voluntarily, which is a matter of right in many states and is rarely denied in federal court. As a result, every bellwether case presents the risk that the plaintiffs will pull the plug if they do not like what they see as they proceed toward trial. There are measures a court can take to avoid this “gaming” of the system, such as allowing the defendants to unilaterally select a replacement for every case that plaintiffs dismiss from the pool. The district court here did not enact that solution or anything comparable. In addition, even when cases are tried to verdict, there is always a way to distinguish them. Just the other day, plaintiffs in a prescription drug case responded to a defense verdict with a statement that the defendant had “gotten away with one” because of evidence unique to that case. The upshot is that courts will continue to set bellwether trials because, after all, cases have to be tried. How to make the process work best, however, will vary from case to case, and the level of success will depend on how the process is implemented.