Litigation Outcomes in State and Federal Courts: a Statistical Portrait

• Publication year: 1996
• Citation: Vol. 19 No. 03

UNIVERSITY OF PUGET SOUND LAW REVIEWVolume 19, No. 3SPRING 1996

Litigation Outcomes in State and Federal Courts: A Statistical Portrait

Theodore Eisenberg, John Goerdt, Brian Ostrom, and David Rottman(fn*)

"U.S. Juries Grow Tougher on Plaintiffs in Lawsuits," the New York Times page-one headline reads.(fn1) The story details how, in 1992, plaintiffs won 52 percent of the personal injury cases decided by jury verdicts, a decline from the 63 percent plaintiff success rate in 1989. The sound-byte explanations follow, including the notion that juries have learned that they, as part of the general population, ultimately pay the costs of high verdicts. Similar stories, reporting both increases and decreases injury award levels, regularly make headlines.(fn2) Jury Verdict Research, Inc. (JVR), a commercial service that sells case outcome information, often is the source of the stories.(fn3)

The stories highlight a major gap in our knowledge of the legal system. Reported aggregate data tend to be exaggerated or incorrect. For example, the figures reported in the Times article almost certainly inflate plaintiff success rates for 1989 and report a time trend that probably does not exist. In an era when court reform and tort reform are constantly on the public policy agenda, the need for accurate national data about the litigation system is more important than ever.

This Article supplies the first comprehensive national assessment of litigation outcomes in state and federal courts. It uses data gathered by the National Center for State Courts and the Administrative Office of the United States Courts. Both data sources are national in scope and derive their information directly from court clerks' offices. The data portray a litigation system with case outcome patterns that differ from the patterns based on less comprehensive sources.

Our principal findings are: (1) plaintiff win rates injury trials in state and federal court are strikingly similar; (2) award levels are much higher in federal court than in state court; (3) federal courts handle a relatively small fraction of the jury trials, but they distribute a surprisingly large percentage of the funds awarded in jury trials; (4) there probably is no significant time trend in plaintiff win rates in federal court jury trials; and (5) cases at almost every stage of disposition proceed more slowly through state courts than through federal courts.

I. The Data

Before reporting our results, it is helpful to describe the data sources on which we rely. We rely on the two most comprehensive sources of data about state and federal case outcomes. The state data come from the Civil Trial Court Network (CTCN), a joint project of the National Center for State Courts (NCSC) and the Bureau of Justice Statistics (BJS), which covers state courts of general jurisdiction in a sample consisting of 45 of the 75 most populous counties in the United States.(fn4) The counties include approximately 33 percent of the 248,709,873 million people reported to be the 1990 United States population.(fn5) The CTCN data cover fiscal 1991-92 (July 1 to June 30) and include a general civil case sample, as well as a jury trial data set.(fn6) The data in the general civil case sample constitute a sample of all tort, contract, and property cases in the 45 sampled counties.(fn7) For each sampled case, the subject area of the case and the mode of disposition (e.g., jury trial) are known.(fn8) The jury trial data set, while limited to jury trials, includes more detailed information about each sampled case, including subject matter area, prevailing party, amount awarded in damages, and time to disposition.(fn9)

The federal data used here, gathered by the Administrative Office of the United States Courts, cover the fiscal years 1979-93.(fn10) When any civil case terminates in federal district court, the court clerk files with the Administrative Office a form containing information about the case. The form includes data regarding the subject matter; the jurisdictional basis; the dates of filing and termination; the procedural progress of the case at termination, the method of disposition, the date a judgment was entered, who prevailed, and the amount awarded in damages. The form distinguishes among many subject matter categories, including branches of tort, contract, and other areas of law.

Because the federal data cover many years, and the state data cover one year, fiscal 1991-92, we sometimes report federal results only for the period that corresponds to the period of the state CTCN data. The federal data for the other years supply a check on whether the 1991-92 federal results are unusual. Since we are primarily interested in comparing results across the common areas of tort and contract law, we limit the federal data to those cases in which diversity of citizenship constitutes the basis of federal jurisdiction. Thus, for all cases reported here, state law governed the case whether the case was adjudicated in state or federal court and whether or not the case could have been brought in state court.

The dual sources of data also raise the issue of whether the subcategories of cases are comparable. In the tables below, we report by subcategory only those classes of cases that are comparable across the state and federal databases: contract, torts, and five subcategories of tort (motor vehicle, other tort, medical malpractice, products liability, and toxic torts). Thus, in the tables, the "All Tort" rows do not reflect a simple sum of the five tort subcategories because we report separately only the tort categories that are comparable across the federal and state data sets. We include cases that do not fit in one of the tort subcategories in the "All Tort" rows but not in any other rows in the tables. Similarly, the "All Cases" rows do not reflect a simple sum of the "All Contract" and "All Tort" rows. The "All Cases" category includes some state cases and federal diversity cases in other areas of law (mostly property law) that are not included in any other rows in the tables. As can be seen from the tables, the vast majority of diversity cases are contract and tort cases.

Differences in the federal and state data sets with respect to two tort subcategories should also be noted. First, the state-based CTCN data employ a residual category, labeled "other tort." The most closely analogous federal case category is "other personal injury." These residual categories differ in their makeup, but each of the two residual categories is the closest of any major category to being a general negligence category. Second, the toxic tort category for the federal data is limited to asbestos cases, the only toxic tort category separately identified in the federal data. In the CTCN data, the toxic tort category includes nonasbestos cases, but asbestos cases dominate the category.(fn11)

II. Plaintiff Win Rates in Jury Trials

We first discuss plaintiff win rates in jury trials. Table 1 reports win rates for the sampled state courts in 1991-92, for all federal district courts in 1991-92, and for all federal district courts for the period 1979-93. For the fiscal year 1991-92, the success rates for similar categories of cases in federal and state courts were strikingly similar. Both the absolute level of success and the relative ranking of categories transcends the state-federal boundary.

In state court contract jury trials in 1991-92, plaintiffs prevailed in 62 percent of the cases versus 60 percent in federal court contract jury trials. In the combined tort category, the success rate in state court was 49 percent versus 55 percent in federal court. In the general residual personal injury tort category, the success rate in state and federal courts was almost equal. In both state and federal courts, noticeably lower success rates exist in the two most discussed areas of modern tort law: products liability and medical malpractice. In medical malpractice jury trials, plaintiff success rates in state court are 30 percent and in federal court are 26 percent. In products liability cases, success rates are 40 percent in state court and 37 percent in federal court. Asbestos cases in both state and federal courts showed the highest win rates at trial. The federal column, reporting results for 1979-93, indicates that the federal results for 1991-92 are atypical largely in their unusually high plaintiff success rate in motor vehicle and products liability cases.

Table 1

Jury Trial Win Rates

Category Name	State N 1991-1992	State Win Rate 1991-1992	Federal N 1991-1992	Federal Win Rate 1991-1992	Federal N 1979-1993	Federal Win Rate 1979-1993
All Contract	2,134	.62	463	.60	8,182	.62
All Tort	9,308	.49	1,124	.55	18,377	.46
Motor Vehicle	3,813	.60	249	.68	4,521	.61
Other Tort	873	.45	304	.46	4,796	.48
Medical Malpractice	1,352	.30	73	.26	1,040	.27
Products Liability	356	.40	218	.37	5,057	.30

...