An Empirical Study of Class Action Settlements and Their Fee Awards


  • Research for this article was supported by Vanderbilt's Cecil D. Branstetter Litigation & Dispute Resolution Program and Law & Business Program. I am grateful for comments I received from Dale Collins, Robin Effron, Ted Eisenberg, Deborah Hensler, Richard Nagareda, Randall Thomas, an anonymous referee for this journal, and participants at workshops at Vanderbilt Law School, the University of Minnesota Law School, the 2009 Meeting of the Midwestern Law and Economics Association, and the 2009 Conference on Empirical Legal Studies. I am also grateful for the research assistance of Drew Dorner, David Dunn, James Gottry, Chris Lantz, Gary Peeples, Keith Randall, Andrew Yi, and, especially, Jessica Pan.

Vanderbilt Law School, 131 21st Ave. S., Nashville, TN 37203; email:


This article is a comprehensive empirical study of class action settlements in federal court. Although there have been prior empirical studies of federal class action settlements, these studies have either been confined to securities cases or have been based on samples of cases that were not intended to be representative of the whole (such as those settlements approved in published opinions). By contrast, in this article, I attempt to study every federal class action settlement from the years 2006 and 2007. As far as I am aware, this study is the first attempt to collect a complete set of federal class action settlements for any given year. I find that district court judges approved 688 class action settlements over this two-year period, involving nearly $33 billion. Of this $33 billion, roughly $5 billion was awarded to class action lawyers, or about 15 percent of the total. Most judges chose to award fees by using the highly discretionary percentage-of-the-settlement method, and the fees awarded according to this method varied over a broad range, with a mean and median around 25 percent. Fee percentages were strongly and inversely associated with the size of the settlement. The age of the case at settlement was positively associated with fee percentages. There was some variation in fee percentages depending on the subject matter of the litigation and the geographic circuit in which the district court was located, with lower percentages in securities cases and in settlements from the Second and Ninth Circuits. There was no evidence that fee percentages were associated with whether the class action was certified as a settlement class or with the political affiliation of the judge who made the award.