Argument preview: Justices to consider propriety of “cy pres” class-action settlements

Wednesday morning the newly constituted bench will consider Frank v. Gaos, the first important class-action case of the term. As most readers of this blog will know, the Supreme Court over the last several years has issued a series of major decisions trimming back the availability of class actions, typically by constraining the broad discretion under which district courts traditionally have supervised those cases. Justice Anthony Kennedy was notable in those cases, often decided by a 5-4 vote, as one of the justices most persistently skeptical of the value of class adjudication. This will be the first opportunity to see if that skepticism persists on the post-Kennedy bench.

The topic for Wednesday is the “cy pres” settlement, a frequent device in class actions that award trivial sums to a large class of plaintiffs. When the costs of identifying class members and distributing proceeds make it impractical to distribute an award to those plaintiffs, courts commonly approve a settlement that instead takes the settlement proceeds and distributes them to some public-interest or charitable recipient that serves an interest parallel to the interests served by the judgment in the case. Those arrangements are called “cy pres” settlements on the theory that they come as close as possible (“cy près comme possible”) to awarding damages to the plaintiffs.

In this case, for example, the plaintiff class alleged that the “headers” that Google transmits when a user clicks on a link in search results generated by Google disseminate private information of the user in violation of the Stored Communications Act. Eventually, Google agreed to a settlement with the named plaintiffs. Google revised various pages on its website to provide disclosure of the information included in the relevant headers. As part of the settlement, Google also agreed to pay damages of $8.5 million. In due course, the district court approved the settlement, as well as a payment of $2.1 million in attorney’s fees to counsel for the plaintiffs. The remainder of the settlement (about $6.5 million) would have provided approximately 6.5¢ to each of the 100 million class members. Concluding that it was not feasible to distribute the fund to the class members (because it would cost more than 6.5¢ per plaintiff to make a distribution), the court instead distributed the settlement fund to initiatives studying internet privacy and involuntary information-sharing at a variety of locations including such institutions as Carnegie Mellon University and the law schools at Harvard, Stanford and Chicago-Kent.

Those settlements have been controversial for numerous reasons, ranging from the poor optics of a class action in which the attorneys are compensated but not the plaintiff class (whose claims are discharged by the settlement), to concerns about the propriety of judges (or counsel) selecting cy pres recipients from institutions with which they are affiliated or from which they graduated. Thus, Congress has considered (but never adopted) legislation that would delineate or curtail the circumstances in which such settlements are permissible. The committees that monitor the Federal Rules of Civil Procedure also have considered the topic in detail but have never recommended changes addressing the subject. And in the Supreme Court, Chief Justice John Roberts issued a statement almost five years ago concurring in the court’s denial of a petition seeking review of a cy pres settlement, but suggesting (in a shot across the bow) that the issue was one that ultimately would require the court’s attention.

On the merits, the briefs make one thing amply clear: There is no rule or statute that provides any substantial guidance to the justices considering the matter. The Federal Rules of Civil Procedure include lots of details about class actions, but they say nothing that fairly can be regarded as authorizing or forbidding the practice. No federal statute addresses the question. And nothing in the Supreme Court’s own decisions offers any relevant guidance. Thus, the justices are almost entirely unconstrained, free to resolve the matter pretty much as they see fit.

The petitioners are a group of class members objecting to the settlement (led by Theodore Frank). Their principal argument is that cy pres settlements create unacceptable conflicts of interest between counsel (who profit from the settlements that pay their attorney’s fees) and the class members (who receive no monetary recovery from a settlement that releases their claims against the defendant). They also emphasize how unseemly it is for the settlements – which by hypothesis cannot practicably be distributed to class members – to be distributed so routinely to institutions (like Harvard and Stanford in this case) that the counsel in the matter attended. Relying on those concerns, petitioners press a hard-line rule that absolutely would bar courts from approving a settlement class if the settlement does not provide direct relief to the class.

In the eccentric procedural posture of the case, the respondents are two groups that support the settlement: the plaintiffs and defendants who agreed to it. Predictably enough, they emphasize not only the broad discretion traditionally afforded district courts to approve settlements but also the features of this particular settlement that might persuade the Supreme Court to view it sympathetically. Most obviously, Rule 23 allows district courts to approve settlements that are “fair, reasonable, and adequate,” a phrase redolent of discretion to consider the unique attributes of each case.

Both groups of respondents emphasize the three limitations that courts have developed to cabin the approval of cy pres settlements: Distribution of proceeds must be infeasible; the cy pres recipients must provide indirect benefits to the class-member interests that the suit pursued; and the recipients must be selected by merit rather than affiliation. Here, for example, it is plain that the proceeds (a few pennies per class member) could not practicably be distributed to class members; there is an apparent link between the public-interest projects receiving the funds and the internet-privacy problems challenged by the plaintiffs; and the award recipients were selected after an open process that permitted submission of proposals from all interested institutions.

To my mind, the case really presents the justices with a problem of allocating institutional responsibility. It seems to me most unlikely that the justices would coalesce around the bright-line rule that Frank suggests (categorically barring all settlements that do not benefit the class). Moreover, it is not even clear that the case fairly presents that question; the respondents contend that Google in fact agreed in the settlement to make permanent changes to its disclosures that exceeded the changes it previously had made voluntarily. The settlement-induced changes at least arguably provide direct prospective relief to the class members, albeit not a single penny of monetary relief.

If the justices get past that point, what confronts them is elucidation of the circumstances in which cy pres settlements are appropriate. Most observers probably would agree that Congress (or the responsible rules committees) are better placed to develop rules or guidelines on that question than the justices, but at this point it is quixotic to expect either group to produce anything substantive on the topic at any point in the foreseeable future. At bottom, then, I expect the case will hinge on whether the unseemly aspects of the developing cy pres practice in the lower courts will motivate the justices to cut back on the relatively unguided inquiries that have confined district courts to date. Presumably the argument will give us more than a glimpse of the sensibilities the justices bring to that question.