The Efficiency of Federal Appellate Decisions: An Examination of Published and Unpublished Opinions

Article excerpt

Almost 60 years ago, James Gough, wrote of the "pathologies" of the federal appellate process-particularly institutional characteristics that obstructed swift, meaningful justice. While some researchers have explored these institutional pathologies for published cases, the present piece extends our understanding in two ways. First, we introduce an empirical measure of judicial efficiency, "swift justice," in terms of adjudicatory disposition time. Second, we examine the impact of particular institutional characteristics-whether a case is orally argued, or whether the decision is published-on judicial efficiency. We analyze the population of published and unpublished U.S. federal appellate decisions from 1976-97, using institutional variables to identify prescriptions that might attenuate pathologies of adjudicatory speed. Our historical analysis provides some baseline evidence that publication status influences the timeliness of appellate decision making.

Until the late 1990s scholars had little knowledge about the institutional dynamics of the U.S. Courts of Appeals. Part of the limitation in scholarly exploration of these courts was the scope and magnitude of collecting data on the thirteen circuits reaching across the country, with different judges, political cultures, and legal influences. The most comprehensive collection of data on these courts was initiated by Donald Songer and funded by the National Science Foundation in 1989. Songer (1996) produced a multiuser data set of approximately 20,000 published decisions from the U.S. Courts of Appeals (hereafter, "Songer Database") that included 229 variables and a sample of published cases decided from 1925 to 1996. In 2003 the National Science Foundation funded an update (hereafter, "Update") to the Songer Database that mirrored Songer's coding scheme and added over 2,000 cases from 1997 to 2002 (Kuersten and Haire, 2007).1

These two rich sources of data have allowed scholars to test a wide variety of substantively meaningful questions surrounding the U.S. Courts of Appeals and other courts, including mixed-outcome decisions (Lindquist, Haire, and Songer, 2007); issue agenda (Songer, Sheehan, and Haire, 2000); party advantage (Clermont and Eisenberg, 2002); the impact of ideology (Hettinger, Lindquist, and Martinek, 2004) and presidential appointment (Giles, Hettinger, and Peppers, 2001); precedent (Cross, 2007); statutes (Randazzo, Waterman, and Fine, 2006); litigant resources (Songer, Sheehan, and Haire, 1999); and judge gender, race (Collins and Moyer, 2008), and tenure (Kaheny, Haire, and Benesh, 2008). The data were also used to produce at least three book-length manuscripts (Cross, 2007; Kuersten and Songer, 2001; Songer, Sheehan, and Haire, 2000). Notwithstanding these contributions, we contend that two areas of necessary appellate research have been relatively neglected.

First, we know little about the institutional consequences of unpublished opinions. The preceding references illustrate that answers to many of the important questions in the subfield of judicial behavior are based on the Songer Database and the Update. Notwithstanding the database's strengths, these data only include opinions published in the Federal Reporter, often called published opinions.2 Unfortunately, as Songer (Songer, 1990; Songer and Sheehan, 1992) himself noted, focusing on reported decisions is highly problematic for several reasons. In the last forty years, a significant portion of the cases decided by the courts of appeals have not been published in the Federal Reporter. Indeed, over time, the number of unreported decisions has started to dwarf the number of those reported; by the 1980s, less than half of the courts' decisions were reported (Songer, 1990), and by 1997 that proportion dropped to one quarter (Mecham, 1997).

But the importance of unpublished decisions has also increased. Federal Rule of Appellate Procedure 32.1 was amended in 2007 to allow attorneys and judges to cite unpublished cases as precedent in all circuits (Boyeskie, 2008; Gant, 2006). …