An Empirical Investigation into Appellate Structure and the Perceived Quality of Appellate Review

Abstract

Commentators have theorized that several factors may improve the process, and thus perhaps the accuracy, of appellate review: (1) review by a panel of judges, (2) subject-matter expertise in the area of the appeal, (3) other lawfinding ability, (4) adherence to traditional notions of appellate hierarchy, and (5) the judicial independence of appellate judges. The considerable discussion that has expounded upon these theories has occurred in a vacuum of abstract generalization. This Article adds a new dimension by presenting the results of an empirical study of bankruptcy appellate opinions issued over a three-year period. The federal bankruptcy appellate structure provides certain litigants the choice to appeal, in the first instance, to one of two distinct appellate tribunals—district courts and bankruptcy appellate panels (BAPs)—whose structural features relating to the theorized qualities of appellate review differ. As BAPs appear to have more of the features identified as improving the quality of appellate review, the study tests the theory through various hypotheses that focus on the perception held by other federal courts within the bankruptcy appellate structure of the quality of appellate review provided by these distinct appellate tribunals. The data show that, as measured by (1) the subsequent disposition rendered by courts of appeals and (2) the citation practices of other federal courts to the appellate opinions issued by BAPs and district courts, BAPs have been perceived to provide a better quality of appellate review. Having unearthed some evidence that supports the theoretical notions underlying the quality of appellate review, this Article concludes that commentators and policymakers ought to be encouraged to explore further, in a more detailed manner, the question of how appellate structure can be designed to produce better results.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 60 citations exist upstream.