The need for a national civil justice survey of incidence and claiming behavior.

• Author: Eisenberg, Theodore
• Position: 2008 ABA Section of Litigation Access to Justice Symposium

Introduction I. Selected Available Civil Justice Data and Their Limitations A. Leading Civil Justice Data Contain Surprises B. Limitations of Existing Civil Justice Data and the Benefits of a National Civil Justice Survey II. What to Track? A. The Nature of the Activities Generating Civil Justice Needs B. The Subject Areas of Civil Justice Activities C. Information to Be Gathered About Civil Justice Incidents and Related Matters III. Prior Research Methodologies and Results on Civil Incidents Conclusion: The Benefits and Plausibility of a National Civil Justice Survey INTRODUCTION

Civil justice issues play a prominent role in society. Family law issues such as divorce (1) and child custody, consumer victimization issues raised by questionable trade practices, (2) and tort issues raised by surprisingly high estimated rates of medical malpractice, (3) questionable prescription drug practices, (4) and other behaviors are part of the fabric of daily life. Policymakers and interest groups regularly debate and assess whether civil problems are best resolved by legislative action, agency action, litigation, alternative dispute resolution, other methods, or some combination of actions. Yet we lack systematic quantitative knowledge about the primary events in daily life that generate civil justice issues. This paper explores the desirability of, and issues related to, creating what I refer to as a national civil justice survey ("NCJS"), analogous to the National Crime Victimization Survey ("NCVS").

The NCVS is the primary source of information on criminal victimization. (5) The survey enables the Bureau of Justice Statistics ("BJS") to estimate the likelihood of many crimes "for the population as a whole as well as for segments of the population such as women, the elderly, members of various racial groups, city dwellers, or other groups." (6) In 2005, U.S. residents age twelve or older experienced about 20 violent crimes per 1,000 people and about 150 property crimes per 1,000 people. (7) In comparison, decades-old national research on incidence of civil problems suggests that adults experience a long-term risk of serious personal injury at the rate of 120 per 1,000 and a risk of serious property damage of 400 per 1,000. (8) A more geographically limited early 1980s survey found that a three-year risk of having a civil justice grievance was 416 per 1,000. (9) The rate of civil justice incidents plainly is high enough to warrant systematic quantitative knowledge of their patterns.

Part I of this Article briefly reviews selected available civil justice data and their limitations. Part II provides a preliminary discussion of the kind of information about civil justice events that might be gathered in a NCJS. Part III reviews methodologies and results in prior civil justice surveys. Part IV briefly suggests the benefits and feasibility of a NCJS.

1. SELECTED AVAILABLE CIVIL JUSTICE DATA AND THEIR LIMITATIONS

   Important and useful civil justice data exist. BJS projects as well as those of other federal agencies supply much of that information. BJS data tend to focus on the end point of the civil disputing process--litigation-and not on the underlying pattern of grievances and claiming behavior that generate observable disputes. (10) Other data sets, for topics like divorce rates and patient safety data, (11) might already provide adequate information about particular topics. (12) In general, however, sources of civil justice data about other topics are sporadic and depend on reporting by intermediaries rather than by those experiencing the problems. Similar to data collection methods used for crime victimization, a household level survey could provide the most reliable information to assess the true extent of civil-justice-related activity. For purposes of this paper, I try to include a reasonably comprehensive list of civil justice topics that might be included in a NCJS. If it is determined that satisfactory information is already systematically gathered about one or more of the topics, the necessary scope of a national civil survey would be reduced accordingly. In order to illustrate the utility of a NCJS, I first focus on how it might enhance the utility of existing BJS data in relation to civil justice.

   1. Leading Civil Justice Data Contain Surprises

      Existing BJS civil justice initiatives have already established their value by providing significant insights into civil justice system performance. The BJS and the National Center for State Courts ("NCSC") make available online and through print reports the best existing information about state courts, including trial outcomes and filings. (13) For example the BJS Report, Civil Trial Cases and Verdicts in Large Counties, reports time trends from 1992 to 2001, in the number of civil trial cases and the amount of jury awards. (14) These data shed light on the operation of our civil justice system, in which the vast majority of cases and trials are adjudicated in state court.

      Some core BJS-NCSC results are truly striking. In 1992, state courts in the nation's seventy-five largest counties are estimated to have concluded 22,451 trials. By 2001, state courts in these counties concluded only 11,908 trials, a reduction of 47.0%. (15) By 2005, the estimated number had fallen to 10,813, a decline from 1992 of 51.8%. (16) The sharpest decreases came in product liability and real property cases, with reductions of 76.0% and 80.1%, respectively, from 1992 to 2001. (17) But by 2005, product liability trials had increased by 42.2% and real property trials had increased by 14.8% since 2001. (18) The BJS-NCSC data, through a methodology consistently applied over the course of fourteen years, thus conclusively established the reduction in trials in state courts. The vanishing trial and its implications for the justice system has been the topic of extended discusdiscussion. (19) With respect to amounts awarded at trial, the results are equally interesting. In 1992, the median jury award in all tort cases, adjusted for inflation to 2005 dollars, was $71,000. (20) In 2001, the median award was only $31,000, a statistically significant decline, followed by a 6.5% increase in 2005 to $33,000. (21) During that same period, awards were down in automobile cases, and up in product liability and medical malpractice cases. (22)

      NCSC data on time trends in case filings, though limited to the subset of states that report information on a consistent basis, are noteworthy as well. For example, Figure 1 shows a long-term decline in tort filings, accompanied by a more modest decline in medical malpractice filings and little pattern in product liability filings: (23)

      [FIGURE 1 OMITTED]

   2. Limitations of Existing Civil Justice Data and the Benefits of a National Civil Justice Survey

      So we have, as exemplified by tort statistics, a downward trend in filings, a downward trend in the number of trials, and a long-term downward trend in median awards. As telling as these data are, we cannot fully know what to make of them because we lack information about the possible number of grievances and disputes underlying them.

      To show why, let's continue to pursue torts as an example. It is one thing if NCSC's declining tort filings are observed in light of a background of a stable rate of tort incidents over time. lt is quite another if declining tort filings are observed and the rate of tort incidents per capita either has substantially increased or decreased. Unless we know about the number of underlying tort incidents, interpreting filings data is subject to unavoidable limitations. Policymakers cannot tell if legislative or other initiatives have had an effect in the expected direction or in an unintended one. It may be that tort reforms that reduce liability exposure increase the number of tort incidents. This would need to be balanced against the presumed litigation savings in order to fully understand outcome patterns. This uncertainty is, of course, equally true of other civil justice subject areas, including consumer problems such as credit card and mortgage disputes.

      Gathering systematic data about the rate of underlying tort and other civil justice incidents over time has other important benefits, though not directly related to case filing and outcome patterns. Estimations of the rate of tort incidents, and the rate at which incidents are satisfactorily resolved, would yield important knowledge about the need for access to civil justice. Specifically, are civil legal services available to those who need them? Are they differentially available based on income, race, gender, or other factors? And how much access to civil justice is in fact needed? A NCJS could provide information beyond that used to note the difficulties in supplying legal services at reasonable costs. By...