Tuesday, May 12, 2009

Finalizing ALI’s Principles Of The Law Of Aggregate Litigation

As members of the American Law Institute (“ALI”) we’ve been closely following the evolution of the Institute’s Principles project on Aggregate Litigation (“PLAL”) – read: class actions – ever since we joined. In prior posts we’ve found plenty about which we were critical. See here, here, here, and here. But for our part we've never given up. We've kept attending meetings, asking questions, and corresponding with the Reporters (those are the law school professors actually responsible for writing the thing) on ways to improve things. And more importantly, we’ve also seen the Reporters make significant efforts to respond to back seat drivers like us. Thus, we’ve acknowledged that the drafts have gotten progressively better and now more closely reflect sensible class action practice in a lot more ways.

That’s true as well with the April 1, 2009 “Proposed Final Draft.”

Aside: This is where we normally link to the place on the ALI’s website where interested persons can purchase a copy of the latest draft – but we’ve looked all around and we don’t see this latest version listed. We don't know why, but the "Final Draft" isn't for sale - at least not yet. [Update: 5/13/09 - the ALI has informed us that the Final Draft is now available for purchase ($45 a pop) at the link we've provided; it was an oversight]

The Final Draft is to be voted on by the ALI’s membership at the Institute’s annual meeting in Washington DC on May 20, 2009. Here’s the meeting agenda. We encourage everyone who’s an ALI member to study the latest draft carefully and to come to the meeting, participate in the debate, and vote.

The world is run by those who show up, folks.

There are a lot of improvements in the Final Draft, but for us the most notable is the elimination of the prior drafts’ attempts (in what was then §2.06) to reduce constitutional constraints upon class actions under the Seventh Amendment. That’s been deleted in the most recent draft, and the Seventh Amendment is acknowledged as a constraint on what courts can do:

In addition, in litigation to which the Seventh Amendment right to jury trial applies, the reconsideration of evidence across the aggregate proceeding and subsequent proceedings must respect the limitations of the Reexamination Clause, which provides that “no fact tried by a jury, shall be otherwise reexamined in any court of the United States, than according to the rules of the common law.”

Final Draft §2.03, comment b. That’s good enough for us. Scratch one major, longstanding objection off the list.

There’s also a more detailed discussion of why class actions are almost always inappropriate in personal injury actions. Such actions “involv[e] high variability and high variation.” Final Draft §2.02, comment b. Even where individual “common issues” might be identified, this variability should preclude class action treatment:
  • “[A]ggregate treatment confined to such a common issue carries a significant risk of an acontextual determination of that issue in isolation from other, related issues.”
  • “[I]ndividual trials additionally would avoid the practical need for evidence on the common issue in an aggregate proceeding to be substantially reconsidered in proceedings on the remaining issues.”
  • “[I]ndividual trials would avoid placing both claimants and respondents at risk of an all-or-nothing determination of the common isssue on the merits in the aggregate.”
  • [T]he predominance-of-the-evidence standard for civil judgments may properly result in some claimants winning and others losing in the early stages of similar claims arising from mass harms."
  • “Multiple individual trials – proceedings that might reach different results – might reflect more accurately the degree of uncertainty associated with a given common issue.”
Id. Thus our initial concerns about the PLAL encouraging class actions in personal injury mass torts (most of what we do) has also been cleared up.

On the other hand, we admit we’re not perfect either. The blatant fraud that’s recently been uncovered in the Nicaraguan banana worker litigation – bogus claims by recruited overseas plaintiffs who used violence and a corrupt “banana republic” court system to conceal their fraud, transcript here (tip of the cap to CalBizLit) – is only the latest in a number of litigation atrocities that American companies have faced while sued for mass torts in foreign courts. Thus, we’ve reconsidered our opposition to PLAL §2.10. That section would make a “consent” procedure available to aggregate litigation by foreign plaintiffs against American companies in American courts. We might even like to see §2.10 strengthened to force foreign plaintiff fraudsters to litigate here if they sue American companies.

But there’s still room for improvement.

On the ALI’s website, there are three amendments that have been offered to PLAL that will be voted upon in the May 20, 2009 session. They concern: (1) single issue classes and preserving the predominance prerequisite to class certification. This amendment was prepared by class action defense guru John Beisner, whose thoughtful comments on the PLAL we’ve previously featured. (2) reducing the emphasis on controversial medical monitoring claims. Bexis put together this amendment. (3) fixing some lingering problems in the choice of law section (§2.05) including the longstanding issue of whether there's a principal-place-of-business rule that would allow plaintiffs all over the country to sue under the law of a single state. This amendment is offered by Vic Schwartz. Note: there are also a couple of other comments posted by the ALI that don't purport to be formal formal amendments. We agree with most of the Morrison comments (which are technical) and reserve judgment on the Weiss comment.

We recommend a “yes” vote on all three amendments, and here’s why.

Proposed Amendment Number 1: Single issue class actions aren’t that frequent and should stay that way. This amendment to PLAL §§2.02 and 2.03 recognizes, first of all, that progress has been made in toning down the earlier drafts’ endorsement of single-issue class actions in all sorts of actions. Compared to what we once thought would be required, these amendments are concise and surgical. For example, there’s no longer anything in the black letter that requires going to the mat. There's no longer anything in §2.02 that overtly champions single-issue class actions. The remaining problems are in the Comments and Reporters’ notes.

Some of this proposed amendment pertains to §2.02 and some to §2.03. We’ll go over the §2.02 aspects first.

The first part of the amendment is, to us, the most important one of all. It adds a simple sentence to §2.02, comment a that recognizes predominance – which is an element of class certification specified in both Fed. R. Civ. P. 23(b)(3) and in the analogous class action rules of virtually every state. This amendment provides that “this Section should not be construed to suggest that the predominance concept does not apply fully to such classes.” The “such” reference is to single issue classes. That's all that's really necessary. Leave predominance alone - precedent demonstrates that it's quite capable of defending itself in court.

Our greatest concern, over the course of all of the prior drafts, has been that the PLAL has sought to weaken the predominance requirement for class certification – which is one element of class certification to which courts have historically given the most “teeth.” Without predominance, courts wouldn’t care nearly as much about things such as reliance in fraud, causation in personal injury cases, exposure in medical monitoring and choice of law generally. Without predominance we’d essentially have drive-by class certification, and that’s not how the law is or how we think it should be. The PLAL's initial outright hostility towards predominance has abated to the point that the Reporters have publicly disavowed any intent to de-emphasize predominance. Thus, we would hope this amendment is not controversial.

The single-issue class amendment includes several corresponding changes (we're not even sure they're technically "amendments") to the reporter’s notes to §2.02. The first, while recognizing that courts have divided on how to treat predominance in single-issue (so-called “(c)(4)”) classes, does not go on to label them “confused,” as the current draft does. We don't think that the split of authority reflects confusion, but rather a genuine difference in approach - whether predominance applies to both the litigation as a whole and to the single issue class, or to just the latter, more narrow, selected issue.

This amendment goes on to equalize the space devoted to both approaches. We think that's still quite generous to the single-issue advocates, since the majority rule (discussed at length in the amendment's rationale section) does not view single-issue classes as an exception to predominance.

The next amendment to the reporters’ notes deals mostly with Canadian law. If' you're wondering what relevance Canadian law has to a principles project devoted to procedure in US courts, join the club (we'd support an amendment striking Canadian law altogether, but there isn't one). The single-issue class amendment moderately contents itself with adding more recent Canadian appellate precedent that disagrees with the Canadian trial decision that’s discussed in the current draft. Thus it recognizes that a similar split of authority exists in Canadian courts as to issue cases (and that's with a Canadian rule that doesn't even contain the US predominance requirement). We agree that, if Canadian precedent is to be used at all, the discussion should be current and complete. That's what this aspect of the amendment does.

The final amendment to the Reporters’ notes for §2.02 recognizes what just about everyone, no matter what position s/he advocates, recognizes is true – single issue classes are not very common. Historically, most courts have denied certification, when plaintiffs have proposed single issue classes. But the current draft PLAL professes to find an “emerging convergence” – based on two Seventh Circuit cases from 2003-04 – supporting single issue classes. That is deleted by this amendment, since any “convergence” (emerging, harmonic, or otherwise) has since remained submerged for another five years. In place of the "convergence" language this amendment includes a selection of cases representing the majority rule limiting single-issue classes, discusses the most frequently proposed separate issue (“general causation”), and points out a fact: that “in most instances” single-issue classes have not succeeded. We agree with that.

The “justification” for this amendment – under good ALI practice substantive amendments must be accompanied by a persuasive legal justification – contains a much more exhaustive treatment of relevant precedent, demonstrating that single-issue classes are often requested but seldom allowed. In addition to Castano (cited in the current draft), the amendment demonstrates that district courts continue to reject single-issue cases (“gerrymandering” one court called it) in a wide variety of contexts. The justification also examines the two Seventh Circuit cases (Mejdrech v. Met-Coil Systems Corp., 319 F.3d 910 (7th Cir. 2003) and Allen v. International Truck & Engine Corp., 358 F.3d 469 (7th Cir. 2004)), and argues that they do not support the broad propositions for which they are cited by the Reporters notes, particularly Allen, which was not a (b)(3) class and thus did not even involve predominance.

The single-issue class amendment lastly jumps to §2.03, and makes one revision to the “Effect on current law” section. The amendment more succinctly states that no rules changes are necessary, and acknowledged the undeniable fact that, at least in product liability and personal injury, most courts reject single-issue classes. Again, this is a significant improvement.

There’s another aspect to the amendment’s justification – practicality. The paucity of single-issue classes over the forty-plus years that Rule 23 has permitted them is equally due to the fact that they just don’t work very well. General causation, for example, doesn’t really prove anything. Almost everything has a dose-response relationship, so the individualized issue of exposure makes this sort of general question essentially meaningless. And almost all diseases have multiple possible causes or vectors, so the “general” question is further confounded by individualized alternative causation issues. Because of this doubly individualized lens through which “general causation” must be focused in any given case, the potential for unconstitutional re-examination of evidence (the Seventh Amendment point) is particularly acute.

From day one, our biggest overall concern about the PLAL as a whole has been its hostility to the predominance requirement and its receptivity to single-issue classes – particularly in areas where such classes have not been allowed by the courts. This simple amendment to the single-issue class discussion goes straight to the heart of that issue and we think goes a long way towards keeping the single-issue-class genie in the bottle.

Now, if we were kings of the class action world, we’d probably do something more drastic – like limit single issue classes to injunctive relief only (รก la Allen) – but we’re not, so we view this amendment as a workable compromise.

Proposed Amendment Number 2: Reduce the emphasis on medical monitoring – since its existence is controversial – and recognize that lots of courts view lots of monitoring schemes as both individualized and divisible.

We reiterate: Bexis drafted and submitted this amendment. We don’t claim to be objective about this (as if we ever do) – we just think it’s right, even considering the source. But we also think it's a good idea to be entirely up front whenever we advocate something here that we're also advocating somewhere else.

Again, this amendment does not affect the black letter law of §2.04. At that level of abstraction, we have no major, amendment-worthy objections. It’s the illustrations that concern us.

As we’ve stated in our prior posts about PLAL, we have two major problems with the medical monitoring discussion contained in the “indivisible” remedies section (now §2.04; formerly §2.05). First, there’s altogether too much attention given to a controversial cause of action that only a minority of the states have adopted and that the ALI itself has never recognized. Not even well-recognized theories warrant eight separate illustrations. Second, the discussion, and especially the order of the illustrations, gives the impression that medical monitoring claims are routinely certifiable as mandatory-non-opt out class actions. That’s just not so. Most medical monitoring claims are not certifiable at all, given individualized exposure and risk issues. And even if they are, a lot of courts don’t consider a lot of proposed medical monitoring to be “indivisible.” Rather plenty of courts view medical monitoring claims as seeking money damages – plus a fig leaf or two.

At the outset, we apologize to the Reporters if they're feeling a little “damned if you do; damned if you don’t.” That is not the intent. We recognize that several of the current illustrations were added to address objections previously made to the questionable substantive status of medical monitoring – such as that a lot of states don't recognize the claim at all (at least not in the absence of present injury).

Those illustrations make valid points, but they aggravated the other problem, that of excessive attention. As a result, we’ve come to the conclusion that the points they make simply shouldn’t be made through illustrations. Rather, illustrations in §2.04 should be used in the same way as illustrations are throughout the PLAL, to illustrate procedural issues concerning aggregation.

So the medical monitoring amendment boils the eight current medical monitoring illustrations down to three. It deletes current Illustration 3, which is when medical monitoring isn’t recognized by substantive law at all, as well as current Illustrations 4, 6, and 9, which address medical monitoring claims where the evidence does not satisfy various substantive elements of a medical monitoring cause of action (testing needs different from the general public, testing not diagnostic, and present injury, respectively). We see those four illustrations as all matters of substantive law, not aggregation.

To replace the current plethora of medical monitoring illustrations, the amendment adds a paragraph stating that medical monitoring is controversial, that the ALI has never voted on whether to recognize it, and that PLAL takes no position on substantive law.

The justification for the amendment contains the most thorough state-by-state breakdown on medical monitoring that we've seen anywhere. The scorecard, as we see it: 11 states, DC and Guam have precedent recognizing medical monitoring; 20 states, federal common law, and the Virgin Islands have precedent rejecting medical monitoring; four states have divergent precedent; and the remaining jurisdictions have no precedent. We thought the 50-state survey was independently valuable, as an overview of medical monitoring, so we’ve already posted it, here.

But now you know that it was actually put together for the ALI. And that’s why we think it’s useful. The ALI commands the best, and we think this is.

The remaining illustrations have been reorganized, and revised reporters notes have been added. As the first illustration, the amendment places what we consider to be the most common result when class certification is sought in a medical monitoring case. That is a denial of certification due to individualized issues such as length/dose of exposure, pre-existing medical risks, contributory fault, etc. The accompanying justification lists literally dozens of cases representative of this illustration. The cases demonstrate that there is no trend away from denial of class certification in medical monitoring cases. Twenty of the cited cases are certification denials from 2003 or later.

We agree that the first illustration involving medical monitorint should reflect the certification decision that is most commonly reached by the courts in that type of case. The amendment does.

The second illustration in the amendment is representative of those cases in which class certification has been granted as to medical monitoring and similar claims. Most of the time, if a medical monitoring class is certified at all, the certification has been under Rule 23(b)(2) – which the PLAL equates with “indivisible” relief. The second illustration in the amendment (which is "Illustration 3"; Illustration 1 is not subject to any amendment and is intact) corresponds to Illustration 2 in the current final draft.

In accordance with the ALI’s tradition of putting client interests at the door (the opposite of how we write this blog), the medical monitoring amendment proposes a fully-researched Reporter’s note to accompany newly rewritten Illustration 2. That note collects more cases than are in the current draft, which have certified non-opt out medical monitoring cases.

In our defense-oriented blog, however, we can’t resist pointing out that those cases tend to be older (only 5 from 2003 or later) and tend to describe things that probably wouldn’t qualify any longer as “medical monitoring.” Even in states that recognize the remedy, medical monitoring is limited to the costs of necessary medical testing. That’s it. Such things as independent studies, data compilation, publication of notices, and the like may or may not be good things – but they aren’t recoverable as medical monitoring. Even where recognized, it's a limited remedy, a snack, not a smorgasbord.

The amendment's third illustration ("Illustration 4") recognizes that resort to a non-opt out class for medical monitoring is, in and of itself, controversial. Demands for medical monitoring have varied quite a bit, and many courts have viewed these demands as thinly disguised damages remedies. The third illustration describes a situation where the court concludes that a proposed remedy masquerading as "medical monitoring" is not, in fact, “indivisible” in terms of the PLAL. The amendment makes clear that this sort of decision is discretionary with the court – involving its evaluation of the wide variety of monitoring requests that can be made.

The accompanying Reporter’s note/justification (both repeat essentially the same thing) collect the cases that have viewed medical monitoring demands as not involving a form of “injunctive” (in terms of the case law) or “indivisible” (in terms of PLAL) relief. There are a lot of them. Thus we agree that this is the third – arguably the second, but the amendment tries to be even-handed – proper illustration concerning class action treatment of medical monitoring claims.

Thus we also support the second, medical monitoring amendment to the PLAL. Of course we do, Bexis wrote it. But no matter who was the author, we would agree that it represents an acceptable fix for the medical monitoring problems that the PLAL has exhibited since the earliest drafts we ever saw.

Proposed Amendment Number 3: Ordinary choice-of-law rules apply in aggregate litigation. This amendment (really a set of amendments, like the first) makes explicit a point that has never been in dispute – that the same choice of law rules should apply in aggregated litigation as in a single case of the same type. That’s a corollary to the principle of the Rules Enabling Act and similar state-law provisions, which is that the creation of procedural rules does not alter any party’s substantive rights. The amendment correctly states that cases purporting to apply different choice-of-law rules in aggregated litigation are contrary to the newly specified point, and it calls out a coupld of those cases.

One aspect of the principle added by the choice-of-law amendment is another ground for rejecting case-law that would apply a defendant’s principal place of business to any and all claims. Cases adopting this rule in aggregated litigation have mostly done so in situations where an individual claim making the same allegations would be subject to a different analysis – usually a First Restatement residence of the injured party/location of the accident rule or a Second Restatement balancing of multiple factors.

We have been critical of prior PLAL drafts because they treat the principal-place-of-business approach to choice of law as if it were on par with other choice-of-law principles. It’s not. Rather that approach has been considered and affirmatively rejected in many cases. We mentioned a few of them here, but there are lots more. It might be a good blog post some day to collect them all, but not today.

Finally, the third aspect of the choice-of-law amendment addresses the all-too-common situation where a court confronts a novel cause of action in aggregated litigation. Usually, well more than half of the implicated jurisdictions haven’t considered the issue at all. The amendment makes clear that resort to the old saw that one “assumes similarity to the forum ‘s law” isn’t enough – particularly in light of constitutional limitations upon applying forum law to out-of-state claims.

An example that strikes home on this blog is the direct-to-consumer advertising exception to the learned intermediary rule, adopted only by New Jersey. Every non-New Jersey case to consider that exception has rejected it, but at best that’s only another ten states. A New Jersey court should not simply assume from the absence of any law that the other 40 or so states would adopt the DTC exception. The opposite assumption is surely closer to the truth.

Similar, albeit less extreme, examples would be medical monitoring (where we’ve already shown that fifteen states haven’t spoken to the issue), public nuisance in product liability, and market share liability. In each of these instances, a novel cause of action has received at best mixed acceptance by the courts. We don’t think that a court in one of the states that has adopted such a theory should be able to take advantage of aggregated litigation to predict that every state yet to speak would do the same.

The more novel the claim, the worse the problem gets – and novel claims have been a hallmark of aggregated litigation.

For these reasons, we’re going to be voting for the third amendment as well.

Finally, we could use the help of all the defense lawyers out there who regularly read our blog. If you're members of ALI, we’d appreciate your considering what we have to say, reviewing the three amendments and the rationale for each, and then coming to the ALI meeting and voting. We know that times are tough and travel budgets lean – but this stuff is important.

Beyond that, whether or not you’re personally an ALI member, if there are ALI members in your firm, could you pass this post (whether you read it online or get it via email) on to them? We’d encourage them to do the same thing: listen to us, study the amendments and their reasoning, and then attend the ALI and vote.

The world is run - and law might be made - by those who show up.

At the beginning of this process, we didn't think it would be possible to improve the PLAL sufficiently that we’d actually consider voting for it. But the Reporters have recently shown more willingness to listen, and we’ve seen a lot of favorable changes. If these three amendments are adopted, we’ll have to reevaluate our position.

No comments: