Though the future of mass torts is uncertain, if the present is any indication, it is bleak. Mass torts have perplexed courts for decades, but only because courts insist on protecting an empty conception of a “day in court.” Consequently, courts have rejected the use of a number of procedures, most notably the class action, to resolve mass tort claims. But the obsession with protecting this conception of a “day in court,” sometimes expressed as a need to protect “litigant autonomy,” or to protect a plaintiff’s “property” in her claim, or to ensure valid “consent” to the resolution of claims, is self-defeating. It leads to more mass torts. To break this vicious circle, courts need to reject the day in court as a procedural ideal.

Mass torts are a consequence of mass production activity in a number of industries, and include torts caused by asbestos and other toxic chemicals, pharmaceuticals, product defects, and oil spills, among many other causes. Mass torts produce a large number of plaintiffs, each with varying circumstances. Many of these plaintiffs, such as those who suffer significant injury or death, have high-value claims. Thus, unlike plaintiffs with small-value claims, most mass tort plaintiffs have sufficient incentive to bring suit separately, and may even value their “autonomy” over their claims. Moreover, resolving mass tort claims en masse may result in inadequate representation, because a subclass of plaintiffs may bias a judgment in its favor, or a class attorney may sell out the plaintiffs for a cheap settlement. Indeed, in two prior cases, Amchem Products, Inc. v. Windsor, 521 U.S. 591 (1997) and Ortiz v. Fibreboard Corp., 527 U.S. 815 (1999), the Supreme Court rejected global settlement class actions in asbestos litigation precisely because of these concerns.

Although these concerns are valid, courts ignore the deterrence produced by mass tort litigation. Many of the industries subject to potential mass tort liability are lightly regulated or, as evidenced by the Minerals Management Service’s performance prior to the BP oil spill, inadequately so. See Ian Urbina, U.S. Said to Allow Drilling Without Needed Permits, N.Y. Times, May 14, 2010, at A1. Consequently, tort liability remains one of the primary methods of regulating these industries, because it deters unlawful behavior.

The deterrent effect of mass tort litigation is far from ideal, however, because the defendant has an inherent advantage over the plaintiffs. Despite the plaintiffs’ varying circumstances, the defendant’s liability will depend on a number of issues that are common to the plaintiffs, such as whether the defendant had prior knowledge of the risks of asbestos exposure. See enkins v. Raymark Indus., 782 F.2d 468, 471-72 (5th Cir. 1986). A defendant will invest more on common issues relative to any individual plaintiff because the defendant has more at stake. A plaintiff seeking $500,000 in damages would balk at spending $500,000 on an epidemiological study to prove causation, and may try to conduct one on the cheap, as the plaintiffs’ attorneys did in the Agent Orange litigation. See Peter H. Shuck, Agent Orange on Trial 53 (1987). But a defendant facing hundreds of millions of dollars in liability will not scrimp on such a study. Thus, a defendant can exploit “naturally occurring economies of scale” to invest in common issues that a plaintiff acting alone cannot. ALI, Principles of the Law of Aggregate Litigation §1.02 cmt. b(3) (2010).

It is true that the plaintiffs can voluntarily aggregate their claims to reach similar economies of scale as the defendant, and such informal aggregation is common in mass tort litigation. But strategic behavior prevents the plaintiffs from completely matching the defendant’s scale advantages. A plaintiff may not want to mix her surefire claim with dubious ones in a class action, or she may simply free-ride off of the work of others. A plaintiff who sues separately, however, shoots herself in the foot. By failing to match the firepower of the defendant, the litigation is not only biased in favor of the defendant, but it weakens the deterrent effect of mass tort liability. A defendant, knowing it can divide and conquer the plaintiffs in any subsequent litigation, will have less incentive to take precautions to avoid mass torts in the first place. Thus, going alone ultimately leads to more mass torts. Few, if anyone, would choose to protect a day in court when the price is cancer or death.

The problem of asymmetric stakes in mass tort litigation is, at bottom, a property problem. The defendant owns all of the liability associated with a common issue, while the recovery (the flip side of liability) is divided among the plaintiffs. Plaintiffs, moreover, cannot aggregate to match the scale of ownership of the defendant primarily because their preferences change after the tort occurs. Before the tort occurs, the plaintiffs would insist on aggregating to deter the defendant and thus prevent the mass tort from happening in the first place. But after the tort occurs, the tort cannot be undone and the plaintiffs only care about recovering as much as possible. This leads to a vicious circle of more mass torts because plaintiffs assert their “litigant autonomy,” which leads to more mass torts, which leads to more “litigant autonomy,” and so on.

Such a precommitment problem is not unique to mass torts, as any smoker or dieter will tell you. Nor is the problem new to the law, as it is analogous to “tragedy of the commons”situations that arise from insufficiently scaled ownership interests. Just as individual grazing rights may cause the overgrazing of commonly owned land, individually owned claims lead to more mass torts—the very thing the claims sought to deter.

To prevent this problem, some form of compelled, collective ownership of the claims is necessary. This can be accomplished through a mandatory class action, which effectively assigns collective ownership of the claims to class counsel. It can also be accomplished through procedures like multidistrict litigation, where a court assigns collective ownership of the claims to lead counsel for common benefit work, at least for pretrial purposes. See Charles Silver & Geoffrey P. Miller, The Quasi-Class Action Method of Managing Multi-District Litigations: Problems and a Proposal, 63 Vand. L. Rev. 107 (2010).

It is true that assigning collective ownership of the claims to a third party may result in the inadequate representation that concerned the Court in Amchem and Ortiz. Inadequate representation, however, is best avoided through other aspects of procedural design that have nothing to do with protecting a day in court. For example, adjustments to the fees paid to class attorneys can ensure that their interests are aligned with those of the class. In fact, a mandatory class action provides much-needed leverage for the plaintiffs in settlement negotiations. Moreover, antisuit injunctions to
stop competing class actions can prevent “reverse auctions” in which the defendant awards a class action settlement to the lowest bidder. Finally, increased use of damage scheduling, or compensation based upon a grid of average awards for certain categories of injuries, can avoid any bias in distributing the recovery. Damage scheduling is perhaps the most controversial of these measures since it may redistribute some of the recovery from high-value to low-value claims. However, averaging already happens when attorneys rely upon prior awards to determine the settlement value of existing cases. See Alexandra D. Lahav, Rough Justice 11-12 (Aug. 9, 2010) (unpublished manuscript), available at http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1562677. It is unclear whether the status quo causes more or less redistribution than damage scheduling.

The resistance to damage scheduling shows just how strangely courts have conceived a day in court. A class action allows for plaintiffs’ ample participation, particularly during the distribution of any recovery, which preserves a meaningful day in court. But the Supreme Court has equated a day in court with an inviolable right to a claim for damages that cannot be taken away without one’s consent. In Ortiz, for example, the Court invoked the “deep-rooted historic tradition that everyone should have his own day in court” to support its rejection of a mandatory settlement class action in asbestos litigation, because the “legal rights of absent class members... are resolved regardless of... their consent.” Ortiz v. Fibreboard Corp., 527 U.S. 815, 846-47 (1999) (internal quotation marks and citation omitted). Recently the Court invoked this same “historic tradition” of a “day in court” to reject the preclusion doctrine of “virtual representation,” noting as an aside that notice is generally required in “class actions seeking monetary relief.” Taylor v. Sturgell, 553 U.S. 880, 892-93, 900-01 n.11 (2008). Finally, in Shady Grove Othopedic Associates, P.A. v. Allstate Insurance Co., Justice Scalia, writing for a plurality, concluded that a class action for statutory damages certified under Federal Rule of Civil Procedure 23 does not “abridge, enlarge, or modify a substantive right,” as required under the Rules Enabling Act, “insofar as it allows willing plaintiffs to join their separate claims against the same defendants in a class action.” 130 S. Ct. 1431, 1443 (2010) (emphasis added). Presumably, in the plurality’s view, a class action that included unwilling plaintiffs would not only “abridge” one’s day in court, but may invalidate Rule 23 in the process.

It does not have to be this way. In its recent decisions, the Court has quoted Hansberry v. Lee for the “principle of general application” that “one is not bound by a judgment in personam in a litigation in which he is not designated as a party.” 311 U.S. 32, 40 (1940). But the Hansberry Court stressed that due process is flexible, and requires only that the procedure, whatever it may be, “fairly insures the protection of the interests of absent parties who are to be bound by it.” Id. at 42. The Court took such a flexible, context-specific approach in Mullane v. Central Hanover Bank & Trust Co., where it examined a New York statute that permitted the aggregation of small trusts for common administration, but allowed for periodic judicial “accountings” which settled, and would preclude, all claims against the common trust fund administrator. 339 U.S. 306, 308-09 (1950). Mullane is generally cited for the proposition that a “fundamental requirement of due process... is notice reasonably calculated... to apprise interested parties,” and indeed the Court criticized the paltry newspaper notice for the settlement proceedings provided by the statute in this case. Id. at 314-15. But the Court permitted such newspaper notice for beneficiaries with unknown “whereabouts,” and noted with respect to unidentifiable, contingent beneficiaries that to require more “would impose a severe burden on the plan, and would likely dissipate its advantages.” Id. at 317-18. Although newspaper notice increased the risk that a beneficiary would lose a day in court, the Court recognized that to protect such a right absolutely may destroy common fund trusts altogether.

Mullane suggests an alternative path from our current trajectory. Instead of fixating on a day in court, courts could follow Mullane by being sensitive to the various interests bound up in the claim, particularly the deterrence it provides. By taking such a context-sensitive approach, courts can recognize that, in the mass tort context, protecting a day in court above all else would “dissipate its advantages.” Thus, to stop the future of mass torts, we may need to look to the past.

Professor Campos offers a provocative argument in favor of mandatory class actions for mass tort litigation. I will attempt to show why, despite his argument’s elegance, its conclusion is flawed. Along the way, I hope to show that neither the present nor the future of mass tort resolution is as bleak as Professor Campos thinks it is. The messy reality of aggregate processing, scattered trials, and mass nonclass settlements lacks the neat symmetry of class litigation, but it strikes a better balance between the need for a level field and the recognition that tort claims belong to tort claimants.

Professor Campos’s argument runs as follows: In multiplaintiff, single-defendant mass litigation, the parties face asymmetric stakes. Therefore defendants rationally spend more resources on the litigation than any plaintiff. This gives defendants an inherent advantage, which in turn reduces overall liability below the ideal level. Inadequate liability means underdeterrence. Without adequate deterrence, potential mass tortfeasors commit more mass torts, causing more harm. The solution, according to Professor Campos, is “compelled, collective ownership of the claims.”

Aggregation is essential for leveling the field in light of the asymmetric stakes that characterize mass tort litigation. To this extent, Professor Campos and I have no disagreement. The question is not whether aggregation is needed, but what form it should take. The form he suggests—mandatory class action—would deprive claimants of control over whether to release their claims in settlement. In order to get the benefits of economies of scale and investment based on aggregate stakes, do plaintiffs really need to relinquish settlement control to representatives and class counsel? No—plaintiffs can obtain most of those benefits by aggregate processing in which steering committees and other leadership counsel perform common benefit work while each plaintiff retains the ultimate right to decide whether to release the claim.

Mass tort litigation in the twenty-first century runs a fairly predictable pattern. Think of major mass torts of the past decade, such as Vioxx, Zyprexa, Baycol, World Trade Center, and the BP oil spill. Each litigation brings its own twists, but the general pattern is consistent: A triggering event occurs. Many lawsuits are filed. Certain plaintiffs’ lawyers emerge as dominant players with large numbers of clients. Federal court cases are transferred for consolidated pretrial proceedings under the Multidistrict Litigation (MDL) statute (or, in the case of the World Trade Center litigation, a statutory forum constraint channels all of the cases to a single federal court, see Air Transportation Safety and System Stabilization Act §408(b)(3)), while some of the state cases are centralized on a statewide basis. Lawyers are appointed to steering committees and other leadership roles. Class certification is denied, at least for the personal injury and wrongful death cases. The cases, therefore, proceed nominally as individual claims, but with lawyers selected to do common benefit work in the MDL performing the lion’s share of the legal work. Court-imposed assessments on individual judgments and settlements compensate these lawyers. A few individual cases may head to trial, often under the aegis of the MDL judge and often with the stated purpose of serving as bellwethers. If the claims have sufficient traction, then at some point settlement negotiations ensue between the defendant and plaintiffs’ leadership counsel. In negotiating a settlement, counsel use the information gleaned from any trials that have occurred, and from discovery conducted in the MDL and any legal rulings by the MDL judge and other judges overseeing consolidated proceedings. If the defendant reaches agreement with the negotiating plaintiffs’ lawyers on the terms of a settlement, the proposed settlement is conveyed to the claimants. Those claimants can decide whether to accept or reject the settlement, and overwhelmingly they accept a settlement their lawyers recommend.

Obviously, this story oversimplifies the course of mass tort litigation and the variations in their lifecycles. In the BP litigation, for example, the defendant established a compensation fund to resolve claims beginning very early in the litigation, even before an MDL judge had been appointed. See Gulf Coast Claims Facility, http://www.gulfcoastclaimsfacility.com (last visited Mar. 4, 2011) . In the Zyrexa litigation, there was a mass settlement before any individual trials occurred. See In re Zyprexa Prods. Liab. Litig., 467 F. Supp. 2d 256, 262-63 (E.D.N.Y. 2006). Multiple defendants and insurers may complicate matters. But for purposes of responding to Professor Campos’s argument, these variations pale in comparison to two critical points about modern mass tort litigation: the endgame is often a wholesale negotiated resolution of claims, and much of the legal work for plaintiffs is handled in aggregated proceedings and performed by lawyers who are well compensated for their common benefit work.

The first point is that mass tort litigation often is resolved through mass settlements, even without class certification. Professor Campos suggests that the obstacle to fully collective resolution of mass torts is an “obsession with protecting a ‘day in court.’” But the “day in court”language misses the point. In mass litigation, no one expects that more than a handful of individual claims will be tried and individually adjudicated. In the Vioxx litigation, fewer than twenty individual claims were tried. See Alexandra D. Lahav, Rediscovering the Social Value of Jurisdictional Redundancy, 82 Tul. L. Rev. 2369, 2394 nn. 106-07 (2008). Merck, however, settled over 33,000 claims in a mass settlement program. See In re Vioxx Prods. Liab. Litig., 650 F. Supp. 2d 549, 559 (E.D. La. 2009). Eli Lilly settled over 25,000 Zyprexa claims in two mass settlements, with zero individual trials. See Alex Berenson, Lilly Settles with 18,000 over Zyprexa, N.Y. Times, Jan. 5, 2007, at C1. The BP oil spill litigation already has resulted in tens of thousands of settlements through the Gulf Coast Claims Facility, with few if any trials on the horizon. Mass tort claims are not resolved in court; they are resolved at the negotiating table by lawyers representing hundreds or thousands of claimants. Better, therefore, to steer clear of the “day in court” language and instead to focus on the real issue: When a defendant offers a mass settlement, do claimants have the right to decline the offer, or can they be compelled by their lawyer and the court to be bound by a particular negotiated resolution of their claims?

The second point is that every mass tort litigation receives aggregated pretrial handling through MDL or comparable forms of consolidation, and well-compensated common-benefit lawyers perform most of the legal work for plaintiffs on discovery, motions, and other pretrial matters. When Professor Campos writes of a defendant’s ability to “divide and conquer the plaintiffs,” he is describing a theoretical problem that does not jibe with the reality of multidistrict litigation and other mass consolidated proceedings. In this regard, it is worth noting that Professor Campos mentions only two actual mass torts: Agent Orange and asbestos. These are important mass torts, to be sure, but they largely reflect the mass tort litigation of the 1980s and the concerns of a prior generation of proceduralists. The question for a new generation should be how to reach just and efficient resolutions in light of the realities of post-Amchem, post-fen-phen mass tort litigation.

In theory, the idea of mandatory class actions offers a way to ensure perfect collectivization and therefore equalization of litigation incentives. In reality, however, what is the marginal gain? Since mass tort litigation, as currently handled, already involves very significant common benefit work by experienced and well-resourced lawyers who are incentivized to invest in the litigation based on aggregate stakes, it is unclear what is actually gained by compelling perfect collectivization through mandatory class actions. Moreover, since mass tort litigation often is resolved through mass aggregate settlements negotiated in the context of MDL and other nonclass litigation, it is unclear how much is gained by compelling a comprehensive resolution.

Professor Campos mentions the possibility of solutions other than mandatory class actions, including multidistrict litigation. His vision of multidistrict litigation, however, bears a troubling resemblance to mandatory class actions. According to Professor Campos, mass torts demand “compelled, collective ownership of the claims,”which can be accomplished by mandatory class actions or by “procedures like multidistrict litigation, where a court assigns collective ownership of the claims to lead counsel for common benefit work, at least for pretrial purposes.” I understand how collective ownership can be a plausible description of the dynamics of class actions (although we should be clear even in the class action setting that the claims belong to the claimants even after class certification). But collective ownership does not and should not describe the role of leadership counsel in MDL and other consolidated nonclass litigation.

Perhaps what Professor Campos characterizes as “collective ownership of the claims” can be better understood in terms of separation of ownership and control. See Principles of the Law of Aggregate Litigation §1.05 Reporters’ Notes, cmt. a (Am. Law Inst. 2010) (drawing a connection between the insights of economic literature on the separation of ownership and control in the corporate context and the problems of aggregate litigation). In a mandatory class action, the owners of the claims (class members) are compelled to relinquish control of their claims to class counsel and class representatives. In a nonmandatory class action, the owners similarly relinquish control unless they opt out. The separation of ownership and control presents agency risks but also provides significant advantages to the claims owners. These advantages, such as the incentive for counsel to invest in the litigation based on aggregate stakes and the ability to pursue claims that would not be viable individually, justify class certification in cases with sufficiently cohesive claims. Nonclass mass litigation involves a separation of ownership and control, but to a different degree than class actions. In multidistrict litigation, the owners of the claims (plaintiffs in cases that are centralized) are compelled to relinquish much of the control over pretrial litigation to counsel appointed to do common benefit work. As a practical matter, control over settlement negotiations often shifts to leadership counsel in MDL as well. But unlike in class actions, plaintiffs in nonclass litigation retain the right to decide whether to accept a particular settlement. There is a big difference between aggregate processing and aggregate resolution. MDL involves collective control of pretrial litigation work and collective efforts at generating negotiated resolutions, but it does not and should not involve compelled collective resolution of claims.

“Compelled, collective ownership,” to use Professor Campos’s phrase, upends the notion that a tort claim belongs to the tort claimant. I have tried to show that such a radical approach offers less than it appears to offer, in light of the collective nature of mass tort practice. Professor Campos considers such a radical approach necessary in part because of his assessment that the future of mass torts, like its present, is “bleak.” But I must ask, bleak compared to what? Compared to individual tort claimants, it is not at all clear that mass tort claimants are worse off. In a world of mass collective representation, mass aggregate proceedings, and mass settlement negotiations, claimants in mass litigation benefit from leverage and economies of scale that are unavailable to individual plaintiffs. It is unsurprising that mass torts often result in substantial compensation for individual claimants, and it is worth asking whether the same results would have been obtained had the claimants been unlucky enough to be injured alone.

I am grateful to have Professor Howard Erichson as an interlocutor. He is not only an expert in mass tort litigation who is well respected by both scholars and practitioners, but also someone who takes great care in addressing those with opposing views. I want to return the favor by affording the same respect to Professor Erichson’s views, and, in doing so, clarify where we disagree.

Professor Erichson is right to say that our disagreement is not about whether aggregate procedures should be used in mass tort litigation, but about what form they should take. Our disagreement, however, reveals a difference in priorities. In a nutshell, we disagree on whether we should “upend[] the notion that a tort claims belongs to the tort claimant.” Professor Erichson believes that this notion of claim ownership should be, and for the most part is, accommodated by existing aggregate procedures for mass tort litigation. However, in my view, claim ownership should not be accommodated at all. As I argued in my Opening Statement, protecting claim ownership undermines the prevention of mass torts. Thus, we should be willing to upend the notion of claim ownership, particularly when it leads to the very mass torts the claim is meant to remedy.

Professor Erichson recognizes the problem of asymmetric stakes in mass tort litigation and the unfair advantage that this problem gives the mass tort defendant. He also acknowledges the loss of deterrence this unfair advantage causes. Professor Erichson, however, questions the “marginal gain” of procedures like mandatory class actions that result in what I called “compelled, collective ownership of the claims.” As Professor Erichson correctly concludes, “compelled, collective ownership of the claims” means compelled assignment of dispositive control over the claims to a third party—such as class counsel—with the plaintiffs remaining as beneficiaries.

However, as Professor Erichson points out, the playing field levels out significantly in practice. Unlike the mass torts of the past, mass tort litigation today is resolved through multidistrict litigation and other mass aggregation procedures that involve a fair amount of separation between ownership and control. For Professor Erichson, it is unclear whether the “perfect collectivization” I prescribe “strikes a better balance between the need for a level playing field and the recognition that tort claims belong to tort claimants.”

I question whether we should strike any such balance. Admittedly, the aggregation procedures that Professor Erichson describes significantly narrow the gap between the defendant and the plaintiffs. However, while the plaintiffs may imperfectly collectivize their claims in order to realize greater leverage and economies of scale, the defendant is already perfectly collectivized. The defendant owns all of the potential liability associated with an issue common to the plaintiffs. The defendant, in fact, is perfectly collectivized regardless of whether the plaintiffs proceed through a class action, through multidistrict litigation, or through individual actions. It is unclear why the defendant is allowed to enjoy “perfect collectivization” while the plaintiffs must settle for less. Indeed, in reaching imperfect collectivization, plaintiffs incur costs that the defendant avoids, such as search costs in building client inventories.

Nevertheless, imperfect collectivization may be sufficient. As Professor Erichson notes, plaintiffs’ attorneys who are assigned to serve as lead counsel, to work on steering committees, or to engage in common benefit work, are all well compensated since they typically receive some portion of any judgments or settlements in the litigation. Thus, substantial aggregation leads to substantial incentives to invest in common benefit work. But again, the defendant’s attorneys are not just substantially incentivized to invest in common issues (the flip side of common benefit work). The defendant’s attorneys are perfectly incentivized because, again, the defendant owns all of the liability associated with any issue common to the class. While the plaintiffs’ imperfect aggregation substantially increases the incentive to invest in common benefit work, such aggregation still cannot match the perfect incentives that the defendant has to invest in common issues.

I would further admit the possibility that perfect incentives to invest in common issues may not be necessary. The investments may have diminishing or discontinuous returns. See Lee Anne Fennell, Slices and Lumps 2-3 (Univ. of Chi. Law School, John M. Olin Law & Econ. Working Paper (2d series) No. 395, 2008) available at http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1106421. Consequently, some amount of aggregation, short of perfection, may be all that is needed. After all, the plaintiffs may find a “smoking gun” with “some minimal level of investment,” and most mass tort litigation arises from such triggering events. Richard A. Nagareda, Mass Torts in a World of Settlement 117-18 (2007). However, it is hard to know the amount of aggregation any given case requires. Like all research and development efforts, in mass tort litigation no one knows how much investment is needed to find a “smoking gun” until it is found. Nevertheless, a sufficient level of investment is highly unlikely given the many investments that can maximize the plaintiff’s recovery, such as legal research or hiring attorneys and experts. Again, unlike the plaintiffs, the defendant does not have to settle for less than perfection. Indeed, to fund these investments, the defendant can take advantage of greater financing options than can the plaintiffs, who, due to restrictions like the law of champerty and maintenance, are limited to debt financing. See Benyamin Appelbaum, Putting Money on Lawsuits, Investors Share in the Payouts, N.Y. Times, Nov. 15, 2010, at A1, available at http://www.nytimes.com/2010/11/15/business/15lawsuit.html.

In sum, I agree with Professor Erichson that, in the real world of mass tort litigation, the playing field is substantially leveled. Nevertheless, in the real world, the defendant still has the upper hand. We sacrifice perfection on the plaintiffs’ side, even though the defendant inherently begins with perfection.

It is reasonable to ask, as Professor Erichson does, what is the “marginal gain” from seeking “perfect collectivization” for the plaintiffs, when perfection risks taking the tort claim away from the claimant. But the better question concerns the marginal cost of imperfect procedures. This lack of perfection, particularly when the defendant already has perfection, does more than lead to a bias in favor of the defendant in the litigation. It weakens the incentives for the defendant to avoid mass torts in the first place.

A recent example can be found in the litigation surrounding the Deepwater Horizon rig explosion and subsequent oil spill. Due to lax regulatory oversight, BP cut corners on safety measures that could have prevented a blowout on the rig. In fact, a Congressional investigation found that the rig’s blowout preventer had a hydraulic leak and a dead battery, which may have led to the explosion. Deepwater Horizon Blowout Preventer “Faulty”—Congress, BBC News, http://news.bbc.co.uk/2/hi/americas/8679090.stm (May 13, 2010). Of course, one cannot undo the past, and further investigations may reveal different causes for the explosion. But it is worth asking whether BP would have used a defective blowout preventer had it known it would face perfectly collectivized plaintiffs in future litigation.

The weakened incentives that arise from imperfect collectivization may result in only a marginal decrease in deterrence. In this situation, people get less deterrence, but at least they can keep their claim. But I doubt anyone would willingly make that trade. The marginal decrease in deterrence means an increased risk of harm, which can result in injuries like death, cancer, and disabilities that cannot be remedied by damages. Moreover, expected recovery would likely be less in imperfect collectivization procedures such as multidistrict litigation because there would be lower investment in common benefit work (and thus a lower probability of recovery) in such procedures, relative to a mandatory class action. Finally, recovery is costlier in procedures like multidistrict litigation given the costs of coordinating with other plaintiffs and attorneyscosts that are avoided by a mandatory class action. Indeed, mandatory class actions can further drive down costs by using procedures such as damage scheduling. Thus, in addition to less deterrence, imperfect collectivization procedures such as multidistrict litigation also lead to lower and costlier expected recoveries. I am at a loss as to why anyone would prefer this in the name of keeping her claim. This preference is akin to preferring a fire extinguisher that not only functions imperfectly, but actually causes more fires.

Admittedly, with perfect collectivization from a mandatory class action comes the possibility of inadequate representation. But protecting against inadequate representation has nothing to do with protecting claim ownership, as I argued in my Opening Statement. More importantly, why should claim ownership stop us from doing the best we can? We should not allow the “good enough” to be the enemy of perfect collectivization.

Professor Erichson’s concern with preserving “the notion that the tort claim belongs to the tort claimant”parallels the Supreme Court’s preoccupation with a day in court. I agree with Professor Erichson that a day in court is rarely a possibility in mass tort litigation. But the Court does not mean “day in court” when it says “day in court.” Instead, and as I argued in the Opening Statement, it means an almost absolute right to own one’s tort claim. Indeed, the Court, like Professor Erichson, wants to avoid “upend[ing] the notion that a tort claim belongs to the tort claimant,” as reflected in recent cases concerning class actions, preclusion law, and the Rules Enabling Act. See, e.g., Shady Grove Orthopedic Assocs. v. Allstate Ins. Co., 130 S. Ct. 1431, 1443 (2010) (Scalia, J., for himself, Roberts, C.J., Thomas, J., and Sotomayor, J.) (concluding that a class action under Rule 23 would not violate the Rules Enabling Act at least to the extent that the class action includes “willing plaintiffs”).

However, the cost of not upending that right is an invasion of a higher-order interest—the right to avoid the tort altogether. One could understand this interest in deterrence as a “liberty”interest—“[the] right to be free from, and to obtain judicial relief for, unjustified intrusions on personal security”—as the Court has suggested in prior cases. Ingraham v. Wright, 430 U.S. 651, 673 (1977) (quoting Meyer v. Nebraska, 262 U.S. 390, 399 (1923)). Perhaps by explicitly recognizing an interest in deterrence as a “liberty” interest, we can more clearly see the cost of protecting the tort claim as a “property” interest.

I want to conclude by identifying an important source of agreement. Professor Erichson’s concern with protecting a tort claimant’s control over the claim is motivated, in part, by a recent proposal concerning aggregate settlements from the American Law Institute’s Principles of the Law of Aggregate Litigation. Section 3.17(b) of the Principles permits clients to agree in advance to accept an aggregate settlement if a majority of the plaintiffs vote in favor of it. By design, section 3.17(b) undermines the consent normally required for a valid settlement because it forces a client to accept an offer, whether she likes it or not, if certain conditions are met. It does so in the name of achieving closure or global peace.

In a recent article, Professor Erichson has criticized “closure” as a legitimate rationale for section 3.17(b) because this rationale undermines the “consent” necessary to legitimate the settlement. See Howard M. Erichson & Benjamin C. Zipursky, Consent Versus Closure, 96 Cornell L. Rev. 265 (2011). I, too, reject closure as a legitimate goal, but not to protect consentrather, to protect deterrence. As should be clear, I am more than willing to undermine claim ownership and consent, but only because I want to prevent mass torts from occurring in the first place. In fact, one danger of a focus on closure is that the defendant may be able to escape the full liability associated with its actions, as happened with the September 11th Victim Compensation Fund, where airlines escaped a significant amount of liability. We may disagree on whether ownership of the tort claim should be protected, but on the issue of global peace, Professor Erichson is on the side of angels.

When thinking about the future of mass torts, we might begin by asking what is unlikely to disappear. With a sense of the unchanging, we can better focus on real change and its opportunities and risks. I propose that the following things are unlikely to disappear in the foreseeable future: (1) mass production, mass marketing, mass transportation, and other conditions that similarly situate large numbers of people; (2) mass harms that occasionally result from wrongful conduct; (3) legal obligations that flow from a sense that justice sometimes requires transferring wealth from perpetrators of such wrongful conduct to their victims; (4) a market for lawyers representing victims of such mass harms; and (5) a shared interest by plaintiffs and defendants in resolving claims by settlement rather than adjudication. In other words, regardless of particular procedures for administering and adjudicating such claims, litigation and settlement of mass torts is not a passing phenomenon.

These points of predictability leave plenty up for grabs. Judicial procedures for coordinating mass litigation have evolved significantly in the past several decades and show no sign of stagnating. Just as judicial procedures change, so does the nature of law practice. Recent years have seen dramatic changes in the way plaintiffs’ lawyers gather clients and coordinate efforts; economic and technological developments will continue to fuel these changes in mass litigation practice. Settlement structures evolve as well. Indeed, mass tort settlement structures appear to be in a particularly experimental stage as lawyers seek ways to accomplish closure within the bounds of legal and ethical constraints.

Professor Campos argues in favor of mandatory class actions for mass torts, but what Professor Campos wants is not really procedural reform, but rather a fundamental change in how our legal system conceives of tort claims in the context of mass harms. In his Closing Statement, Professor Campos takes issue with my assertion that tort claims belong to tort claimants. He correctly reads me to believe that individual claim ownership can be accommodated by aggregate procedures in which mass tort litigation is practiced and processed on a collective basis but in which claimants ultimately retain the right to decide whether to release their claims in settlement. Professor Campos responds to this view not with any disagreement about the possibility of such accommodation, but with the argument that “claim ownership should not be accommodated at all.” In his view, we should remove control of tort claims from the claimants because “protecting claim ownership undermines the prevention of mass torts.”

Prevention of mass harms is a worthy goal, but Professor Campos wrongly assumes that embracing this goal warrants reimagining tort claims as collective property. His argument suffers from several defects. First, in focusing on the regulatory role of tort litigation, he loses sight of tort law’s other functions. Second, he displays an overoptimistic faith in the power of tort litigation to alter corporate conduct. Third, he downplays the agency risks in class actions as well as the differences among claimants in personal injury mass torts.

While few would disagree that deterrence is an important goal of tort litigation, Professor Campos assumes either that deterrence is the only worthwhile goal or that it should trump all other goals. Tort litigation functions as an important supplement to direct regulation and Professor Campos is correct that mass harms often reflect gaps and inadequacies in our regulatory systems. In the United States, political resistance to expansive and expensive government bureaucracies means that we rely heavily on litigation to supplement regulatory enforcement. Professor Campos takes this point too far, however, when he suggests that tort claimants’ ownership of claims does not matter. Tort law does not merely impose penalties on wrongdoers—it grants a right of compensation to victims. If ownership of claims is irrelevant, then why should we address the problem of mass harms through tort litigation rather than through other forms of regulation?

Professor Campos exhibits a strong faith not only that litigation can influence corporate conduct for the better, but also that mandatory class actions would appreciably augment this influence as compared with other forms of aggregate litigation. Professor Campos says that “it is worth asking whether BP would have used a defective blowout preventer had it known it would face perfectly collectivized plaintiffs in future litigation.” Admittedly, the impact of procedural reform on any particular piece of corporate behavior is a matter of speculation, but given the extent to which corporations already fear mass litigation, it strikes me as naïve to suggest that the difference between current forms of aggregation and mandatory class actions would have saved the Gulf.

If Professor Campos seems overconfident about class actions’positive influence, perhaps it is because he downplays the risks of class actions. His argument assumes that for any given set of meritorious claims, a class action would result in greater total liability than nonclass mass litigation. Because of agency risks in class actions, the opposite may be true. By strengthening counsel’s control over settlement, class certification creates a risk that class counsel will agree to settle claims for less than their full value. In settlement class actions, counsel may agree to a low settlement price to win the opportunity to move forward as putative class counsel; this is the reverse auction problem Professor John C. Coffee, Jr. identified in Class Wars: The Dilemma of the Mass Tort Class Action, 95 Colum. L. Rev. 1343, 1370 (1995), and the Supreme Court considered in its rejection of an asbestos settlement class action in Amchem Products v. Windsor, 521 U.S. 591 (1997). In litigation class actions, counsel may agree to a low settlement price to guarantee a substantial payday rather than face the risk and expense of protracted litigation. Mandatory class actions exacerbate this risk by removing the possibility of exit. Professor Campos’s deterrence-based argument in favor of mandatory class actions relies entirely on the prospect that such actions would increase defendants’ liability and thereby create a stronger disincentive to harmful conduct, but the settlement dynamics of class actions make this a questionable prospect outside of the context of small-claims class actions.

Courts generally deny class certification in personal injury mass tort cases. See, e.g., In re Vioxx Prods. Liab. Litig., 239 F.R.D. 450 (E.D. La. 2006); In re Baycol Prods. Liab. Litig., 218 F.R.D. 197 (D. Minn. 2003). In light of the individualized issues in personal injury litigation, courts sensibly resist delivering control to class counsel. Without sufficiently cohesive claims, class certification presents not only the lawyer-client conflicts the previous paragraph describes, but also a risk of intraclass conflicts—that is, the danger that groups or individuals within a class may be treated unfairly vis-à-vis the rest of the class. The benefits of class certification require a strong degree of class cohesion. Such cohesion may occur in many types of litigation, including certain consumer claims of economic harm based on uniform corporate conduct. Personal injury claims based on product liability or other mass tortious conduct, however, tend to involve differences among claimants that render class certification both less useful and more dangerous.

Let me be clear that in arguing against mandatory class actions for mass torts, I am not advocating an anti-aggregation position. Aggregate procedures such as multidistrict litigation and statewide consolidation remain essential tools for the efficient processing of claims in mass litigation, and mass collective representation by plaintiffs’ counsel gives plaintiffs needed leverage to level the field with defendants. Rather, I am arguing against a mindset that when mass harms are at issue, individual claimants do not matter.

The Supreme Court understands the class action rule to be “procedural” in the Erie context, see Shady Grove Orthopedic Assocs. v. Allstate Ins. Co., 139 S. Ct. 1431 (2010), and Federal Rule of Civil Procedure 23 has not been held to violate the Rules Enabling Act’s command that rules of procedure “shall not abridge, enlarge or modify any substantive right.” 28 U.S.C. § 2072(b) (2006). When Professor Campos argues for mandatory class actions because “claim ownership should not be accommodated at all,” his reasoning makes the leap from procedure to substance—not in the sense of altering substantive tort law as it applies to each claim, but in the sense that he reconceives the nature of tort claims as collective rather than individual.

There is nothing inherently wrong with altering substantive law in the mass harm setting. Courts have done so, albeit rarely, to reach more accurate liability determinations in cases involving fungible products and multiple defendants. See, e.g., Sindell v. Abbott Laboratories, 607 P.2d 924 (Cal. 1980). One can imagine a similar move to impose proportionate liability in cases involving indeterminate plaintiffs. Professor Campos, however, does not address such reform of substantive tort law as it might apply to each plaintiff’s claim. Rather, whatever the applicable rules of tort liability, Professor Campos seeks to collectivize control of the tort claims of individual victims of mass harm.

In the end, our disagreement is not about whether we favor collective litigation—we both do—but rather about whether at the end of the day the individual tort claimant should have the power to decide whether to release a claim in settlement. The market already creates powerful incentives for collective representation on the plaintiffs’ side, and the judicial system powerfully facilitates collective representation by the appointment of leadership counsel in aggregated litigation. Therefore, even in a world without mandatory mass tort class actions, plaintiffs receive substantial benefits of collective representation. Collective representation creates leverage that drives mass tort settlements. The back-end right of individual claimants to decide whether to accept a settlement incentivizes counsel to ensure an adequate overall settlement and a fair allocation. Professor Campos considers such market-driven and court-facilitated collectivization insufficient; he would have the law impose an absolute form of collectivization. Such absolute collectivization, I have tried to show, carries real dangers but only illusory benefits.