The Class Action Coalition releases a refutation of the much-hyped Eisenberg/Miller study (Jan. 16) to be published in the forthcoming Journal of Empirical Legal Studies purporting to analyze trends in attorneys' fees in federal and state class actions. In an analysis of published opinions, Eisenberg and Miller claim that attorneys' fees in class actions have been stable over time, and that state courts have not been more generous than federal courts in such cases.
But, say the Class Action Coalition, the methodology of the study biases the result; by relying solely on published opinions, the study omits the numerous unpublished settlements. Moreover, as the authors acknowledge, the mix of federal settlements (where securities cases tend to be litigated) is different than state settlements under current law. By averaging all state courts, the paper ignores the fundamental problem of magnet jurisdictions. Finally, the study makes no effort to distinguish between announced and actual relief to the class: a $20 coupon is treated as equivalent to a $20 check. "In short, in state court cases, the relief actually recovered by class members is often far less than what was 'advertised' in the settlement proposal. The authors essentially assume away this problem -- the key problem with state court class action settlements."
The paper was debated at an AEI event today.
[cross-posted from Overlawyered, where it ran Feb. 20, 2004]