| ,

The Negotiation Class

by , and

Vol. 104 No. 1 (2020) | A Clearer View | Download PDF Version of Article
the negotiation class

Growing dockets have long been the mother of judicial invention. In 1968, Congress created the Judicial Panel on Multidistrict Litigation and authorized it to create multidistrict litigations (or MDLs) to consolidate and coordinate similar claims filed in numerous districts across the country. Now, according to recent research by the Federal Judicial Center, MDL actions encompass as much as 21 percent of all private civil filings, up from 5 percent in 2003.

The increase has led to calls for official reform and has been a creative catalyst for those in the practice looking to address the influx more quickly. Opioid lawsuits have set the ultimate stage for imaginative lawyering. The National Prescription Opiate Litigation (MDL 2804) consolidated more than 2,000 pending lawsuits brought by thousands of the nation’s cities, counties, tribal authorities, and individuals against hundreds of manufacturers, marketers, and distributors of opioids, seeking to recover the costs of fighting the epidemic. In addition to the 2,000-plus cases in the MDL, many other cities and counties have filed nonremovable cases in state courts, and most of the 50 state governments have filed cases in state courts. By most accounts, the opioid MDL is the most complex case in U.S. history. And, characteristically, the late Duke Law Professor Francis McGovern, one of three appointed special masters in the opioid MDL, responded with a new approach. In traditional class actions, classes are certified for purposes of litigation and, more recently, for settlement. For the opioid MDL, McGovern proposed the idea of a “negotiation class action.” The court would certify a class for purposes of negotiation, after the plaintiffs had agreed ex ante how to divide any later-achieved lump-sum settlement. If settlement is reached, those class members who did not opt out up front would have to approve it by a supermajority vote.

McGovern had enlisted Harvard Law Professor William Rubenstein, an expert on class actions, to serve as the court’s expert consultant on class-certification issues in the opioid MDL, and the duo drafted a law review article explaining the new idea [The Negotiation Class: A Cooperative Approach to Class Actions Involving Large Stakeholders, 99 TEX. L. REV. (forthcoming 2020)]. Apparently convinced of the approach’s utility and doctrinal soundness, the opioid MDL plaintiffs’ leadership team moved for certification of a negotiation class in the summer of 2019. After highly adversarial proceedings, Judge Dan A. Polster of the U.S. District Court for the Northern District of Ohio gave the green light in a 40-page slip opinion, In re Nat’l Prescription Opiate Litig. At press time, the viability of the negotiation class was pending before the Sixth Circuit on interlocutory appeal. Elizabeth Burch, a University of Georgia law professor and author of Mass Tort Deals: Backroom Bargaining in Multidistrict Litigation (Cambridge University Press, 2019), discussed the implications of the negotiation class with McGovern and Rubenstein in early 2020, just weeks before McGovern’s unexpected death on Feb. 14. (See a tribute to him.) Their conversation follows — and helps to illustrate how much we’ve lost with McGovern’s untimely passing.


Burch: How did you come up with this idea for the negotiation class, and how does it work?

McGovern: The genesis was a phone call from some lawyers in the opioid MDL who were wondering how they could participate in some type of overall settlement, even though they were not part of the Plaintiffs Executive Committee. There are roughly 30,000 cities and counties that would potentially be involved in settlement, so the question was, “How do you put together a group of folks who could facilitate their bargaining power and still provide defendants the kind of closure they are looking for in the settlement of a case?”

Under the negotiation class approach, the plaintiffs agree in advance — prior to any negotiation — how they would divide up the money, and they agree to be bound by a supermajority vote.

Rubenstein: Traditionally, there are two types of class actions. The first is the conventional “trial class action” — where you certify a class and enable people to opt out up front and then go ahead and try the case. Whoever doesn’t opt out is bound to the outcome of the case. The second is a “settlement class action” — where the case is not certified first; the parties settle the case and then simultaneously seek judicial approval of the settlement and class certification.

What’s new about this “negotiation class” idea is that it works like an amalgam of those two. As in the trial class action, we certify the class up front; but as in the settlement class action, we only do so for the purposes of negotiating a settlement. What Francis came up with is: “We can’t tell you exactly how much you’re going to get. But we can tell you what percentage of the settlement you’ll get. And we can safeguard your rights by making sure when a settlement is finally achieved every class member will get to vote on the settlement.” It really is a kind of ingenious idea that puts together a lot of pieces that both enable settlements in certain types of situations and simultaneously protect the rights of absent class members.

Burch: To certify a class under the “negotiation class” proposal, is it enough that plaintiffs would be seeking a common issue of process — that is, to establish a procedural mechanism that provides a voting process as to settlement considerations — or must plaintiffs satisfy Rule 23’s standards by identifying a common question of conduct or liability that exists independent of class certification?

Rubenstein: From a doctrinal perspective, we were both very mindful that, in Amchem Products, Inc. v. Windsor [521 U.S. 591 (1997)], the Supreme Court ruled that a class proposed for settlement purposes still had to meet all of the requirements of class certification (except the manageability requirement). The proposed class even had to face a heightened review of the adequacy of class representatives if they’d already settled the case. And we never imagined deviating from that. So we have proposed that all the class certification requirements would have to be met for the negotiation class to go forward. And as it was applied in the opioid case, Judge Polster’s opinion carefully worked through all of the class certification requirements, which were hotly contested by the adversaries in the case. [See In re Nat’l Prescription Opiate Litig., 332 F.R.D. 532 (N.D. Ohio 2019).]

The Court also said in Amchem that we should be careful about the adequacy of class representatives in these circumstances. Indeed, one of the improvements in the negotiation class certification is that the court is scrutinizing who is representing the class prior to the settlement negotiations. And unlike a settlement class action, where you just get the deal put on the table at the end, here the court examines up front whether the deal will be fair to everyone in the class, and whether the class is adequately represented. So from a doctrinal point of view, I think this is actually an improvement on the settlement class action.

Burch: Are judges at a disadvantage in terms of assessing conflicts of interest at this early stage? If defendants are the primary means forcing information out of the plaintiffs that might bring to light structural conflicts of interests, might plaintiffs be less likely to produce that information on the front end?

Rubenstein: One concern of both settlement class actions and negotiation class actions is that both the plaintiffs and defendants, jointly, are proposing something to the judge and everyone’s saying how great it is. And therefore the judge might not get the kind of information she would in an adversarial-type proceeding. (Though, just as a footnote, it so happened in the opioid case that there was significant opposition to the negotiation class, so Judge Polster received a lot of information for and against the idea.)

However, notice two things about our proposal that I think are better than a settlement class action — or at least different. One, the court will get more information up front because there’s an opt-out period. There’s some opportunity for the class members either to say to the court, “We got notice of this and we don’t like it. It doesn’t smell right.” Two, because the distributional mechanism has to be worked out up front, the court will be able to ensure that the settlement’s apportionment is fair to all the class members before their counsel begin negotiating a settlement for them.

In the opioid case, Judge Polster didn’t wait until settlement time to look at the proposed apportionment. Instead, he appointed a neutral special master to write a report on whether the settlement apportionment seemed fair to all the class members up front.

Burch: What sort of gamesmanship might judges expect to occur (and correspondingly, seek to thwart or exploit) under a negotiation class?

McGovern: The inventiveness of lawyers, I think, has very little limits. As a result, it would be extremely hubristic of us to project that there couldn’t be gaming associated with any kind of new procedural device. So far, when we’ve looked at this within the context of the opioid case, we haven’t seen gamesmanship as much on the plaintiffs’ side of the coin as we have seen on the defendants’ side of the coin. It remains to be seen if that will be the case in other contexts.

We certainly need to think about the limits of this type of approach. I feel very comfortable about its use in the context of the opioid case. But I’m sure, if given the time, I could think of some abuses that might occur in other situations. And that’s why I think having the judge play such an important role at the beginning of the negotiations is so critical. It allows the judge to be confident that any gaming is within the bounds of acceptability. So, sure, the negotiation class approach could be gamed. But I think there are some safeguards there that would, in the hands of a good judge, prevent any unacceptable gaming.

Rubenstein: We see this proposal working in a particular type of collective action situation — one like the opioid case, for instance, where you have some members with enormous stakes in the controversy and other members with smaller stakes. Usually, you would worry whether, in an aggregate settlement, those with the largest stake will have an inordinate role and apportion too much of the money towards themselves. We don’t want them to have a disproportionate impact, but you do want them to stay in the class so that there can be an aggregate settlement.

Fortunately, you have two countervailing aspects to this potential problem, at least in the opioid case. First, the class members with smaller value claims outnumber the class members with significant dollar claims, so on a “one class member-one vote basis,” the full class can outvote the larger stakeholders; this ensures that any distribution plan cannot favor the large stakeholders too much. Second, the court has to look at the distributional plan upfront. Here, public health experts developed the mechanism for dividing up the money, and it is based purely on the opioid epidemic’s effect on each county or city. And to me, examining the legitimacy of the allocation tool is key.

McGovern: The normal negotiation process in mass claims cases often results in the empowerment of certain subgroups more than others. Certainly, much ink has been spilled on the kinds of outcomes that can occur because of such unequal bargaining power. But the concept of the negotiation class is that the bargaining power is collective. In the opioid litigation, we haven’t seen opt-outs from the larger entities that would traditionally have more bargaining power in any kind of negotiation associated with cities and counties. That leveling effect, in terms of bargaining power, is quite astonishing to me.

Burch: The article seems to prioritize compensation and finality over other (perhaps competing) litigation aims such as deterring wrongdoers, empowering victims, generating public goods by unearthing information and making it available to all regulators, and allowing litigants and the public to participate in trials. Is that correct? If so, should the negotiation class be used only in a narrow class of cases in which compensation is the predominant or sole priority? Or are there ways to accommodate competing ends, say, among attorneys general?

McGovern: The practicalities of the situation would determine the answer. If the competing goals are sufficiently strident, then I doubt you would have a successful negotiation class. The only way this approach can be successful, almost by definition, is by avoiding a large number of opt-outs. Implicit in choosing not to opt out is an agreement with some of the fundamental goals of the litigation, the allocation of the money, and the decision making via supermajority vote.

Rubenstein: In the opioid case, for example, it’s fair to say that the class members have expressed different ideas about what the litigation is about. But the opt-outs represent just over 10 percent of the U.S. population. So it appears, at least in this circumstance, that notwithstanding disparate goals among the class members, the mechanism appears to have achieved quite a broad consensus of entities to be involved in it.

But structurally, we tend to think this idea is going to be best in a particular type of situation, where you have the risk of large class members defecting and, relatedly, the risk of being unable to put together a deal that the defendants will agree to. In that context, you are dealing with a negotiation of a monetary settlement that has the usual attendant risks of opt-out, hold-out, etc. A primary goal of the negotiation class is to encourage the class members to cooperate with one another, speak to one another, and interact with one another. So it’s possible for plaintiffs to discuss having different goals, or different ideas, or different thoughts about what to do, and — instead of going off in different directions — come up with a collective solution. That would give them an enormous amount of leverage against the defendant.

Burch: At what point during the litigation process should parties anticipate certifying a negotiation class?

McGovern: That would depend. There is always a risk that you could create a negotiation class prematurely, which might result in an outcome that would not be as beneficial to the members of the class as it would be if you waited until you got more information. But there are a couple of barriers to that happening. One, of course, is the court. Another is the individual members of the class who decide that they need more information before they feel comfortable with a negotiated amount, and their counsel. And another is the class action counsel representing the class. So the risk is there, but there are safeguards as well.

Burch: What sort of guardrails, limits, or barriers would you impose on the use of the negotiation class going forward?

McGovern: Bill actually raised what I worry about the most: Might the negotiation class entity be too powerful? We know from game theory and various literature that working together in groups tends to be superior to working individually. And so a judge may need to make sure that, in creating a bargaining entity through the negotiation class approach, it is not skewing the negotiation balance in an unproductive way. But more generally, I think the ultimate answer to this concern is this: Defendants don’t have to do a deal with the negotiation class. It’s up to them. In fact, I think some of the opposition to the negotiation class stems from the fact that there are alternative approaches to resolving cases that defendants might prefer. I’ll leave it to others to decide whether those alternative approaches give defendants more or less bargaining power.

Rubenstein: We think the class certification requirements still have to be met for negotiation class certification. So that’s an important first guardrail. Second, up-front consideration of whether the allocation of the settlement is equitable is another guardrail. As Judge Polster has pointed out, you wouldn’t want to start up this whole process, get a settlement out of it, have it put to a vote, have the vote approved, and then have the judge find, at the end of the day, that the allocation wasn’t equitable. [See Fed. R. Civ. P. 23(e)(2)(D) (requiring a court reviewing a proposed class action settlement to ensure that “the proposal treats class members equitably relative to each other”).] It would be just a complete waste of time. Third, the negotiation class requires judicial approval of who is negotiating on behalf of this class. From a historical perspective, the reason people didn’t like the settlement class action in the 1970s and the 1980s was that they were shocked that someone could negotiate a classwide settlement who hadn’t been approved by a court to do so. So, again, the negotiation class actually improves on the guardrails of the settlement class by reviewing the class representatives’ and class counsel’s adequacy prior to a negotiation. As a final guardrail, the class itself gets to vote on the outcome — which is an attribute we don’t currently have in any form, other than the negative form of an opt-out. If anything, the way I view this is it actually adds a lot of bells and whistles to the current settlement class action practice.

McGovern: It’s a topic that deserves a lot of additional scrutiny. Like most of my ideas, it comes out of a concrete case. Sometimes the ideas work in one context and not in another. And sometimes they don’t work at all. So it gives me some pleasure to even think that others are discussing this in various places around the country.

Burch: The Sixth Circuit has the case on interlocutory appeal now. Any predictions?

McGovern: There are quite a few judges who view the MDL process as a crucible for inventiveness. They are willing to allow for the idea that complex MDL cases sometimes deserve different kinds of approaches, simply because they are, to a certain extent, idiosyncratic. It would not surprise me to see the Sixth Circuit view the negotiation class in a positive, or, at least, agnostic way, and await what happens next.

It’s also possible that defendants will want to use the negotiation class for the assurance of being able to bargain with a coherent group, and then use a different, more conventional device for resolution. So there may be some value in the establishment of a bargaining entity, even if it is not used as fully contemplated. For that reason, you might see a greater willingness to experiment. One of the problems, always, with any type of novel idea, is the time lag between the invention and ultimate judicial scrutiny.

Rubenstein: I guess only time will tell. I think we all agree it’s going to be interesting — both in terms of what the courts say about our new proposal and, more importantly, in terms of what happens in attempting to generate a meaningful settlement in the case. Ultimately, all of our procedural efforts are just an attempt to help bring a little bit of justice in the world. And there are a lot of important aspects of the case that I think we all hope will get ironed out in a meaningful way soon.