Voir Dire, Vol. 5, Issue 3, Summer 1998, pp. 19-21 & 25, American Board of Trial Advocates.
Setting aside the one case in 100,000 that makes headlines, are juries generally capricious and liberal? Are verdicts frequently unreasonable? In order to effectively assess the quality of the civil jury system, there is a compelling need for research on the quality of jury decision-making, rather than anecdotal, if spectacular, cases.
Whatever the solution, contemplating altering the rules of the civil jury system ought to be based on a careful examination of the outcomes generated by the jury decision-making process, rather than on anecdotal, emotionally manipulative stories of supposed jury incompetence.
My own research, in fact, provides strong evidence that juries are rational, restrained, and thoughtful; that their verdicts are quite predictable. The awards they determine can be mostly anticipated and are based on factors that juries should be considering to determine awards. If anything, juries err on the side of the defendant, not the plaintiff.
Last year, I completed a doctorate in economics at the University of Chicago, for which I wrote a thesis based on an empirical examination of the civil jury system. I was fortunate to have my research supervised by several of the world's premier economists, including the 1992 Nobel Laureate in Economics.1
In my dissertation, I analyzed compensatory and punitive damages awarded by juries for claims arising from injuries sustained in automobile accident cases involving allegations of driving under the influence of alcohol. My objective was to determine if juries were making rational, well-thought-out decisions in awarding compensatory damages. In this article, I give a synopsis of the results of my research and the relevance of these findings to tort reform efforts.
Previous Research on Jury Decision-Making
Ideally, to assess how juries make decisions, we would like to observe their actual behavior in the jury room. Unfortunately, the jury deliberation process is hidden from view. As a matter of fact, for a short time in the mid-1950s, researchers at the University of Chicago were permitted to listen in on the deliberations of several federal juries, but a subsequent Congressional inquiry soon led to legislation prohibiting this.
Thus, exactly how juries reach decisions inside the jury room cannot be observed or determined directly, which presents a thorny research problem: we can observe what is presented to a jury, and we can observe the results of the decisions that juries make, but actual deliberations are hidden to researchers.
We can only infer what factors they base their decisions upon. But, by using econometric analysis, powerful conclusions can be reached based on standard statistical research methodologies. Among the prior research papers published in this area, there are two notable studies that were highly informative and report findings consistent with that of other jury verdict research.
An economist at Harvard University, Kip Viscusi, examined over 11,000 insurance claims and evaluated all claims that paid for pain and suffering.2 The results of his research led him to reject the idea that payments for pain and suffering awards are arbitrary and capricious. In fact, the average payment was rather modest, amounting to approximately $18,000 in 1986 dollars.
Viscusi argues that proposed limits or caps on awards for pain and suffering would negatively impact the few victims of catastrophic injuries, such as brain injury and quadriplegia, while leaving the great majority of awards for lesser injuries unaffected. He found that larger claims for pain and suffering do not receive proportionally larger awards.
A researcher at the RAND Corporation, Robert MacCoun, reviewed a large body of jury verdicts and observed, among other things, that fewer than 9 percent of cases involve punitive awards, and the median in 15 of 20 jurisdictions was below $40,000.3
He also found that over 50 percent of punitive awards were reduced or eliminated in post-trial proceedings, that wealthy defendants do not have a different outcome than other non- institutional defendants, and that while juries tend to award more money than judges, judges tend to find for the plaintiffs more often. MacCoun concluded that there was no evidence to suggest that jurors are less competent than judges as fact finders and cites evidence for advantages in the performance of heterogeneous groups over that of individuals.
What do juries count as important? How much of their decisions can be explained by observable factors? These are important questions to be addressed in assessing the quality of the civil jury system. They are questions I sought to answer in my own research.
A Model of the Jury Decision
In a typical (non-fatal) personal injury case, plaintiffs make claims in several categories for compensation resulting from losses: first, past and future lost earnings; second, household service losses; third, medical and property losses; and fourth, pain and suffering, including loss of enjoyment of life.
These four categories and the factors that affect them are generally observable in every trial and are available, to varying degrees, in the Jury Verdict Research (JVR) data. Claims for compensation can include any one or more of these categories. The job of the jury is to determine the extent of losses in each category, based on the evidence provided by the plaintiff and the defense.
In my statistical analysis I take into consideration these factors through the use of multiple regression, a statistical method which allows me to measure the separate effect of many possible factors that impact the jury's decision. In the process of determining these effects, I can also estimate what portion of the award that cannot be explained by the factors-a way of measuring how "rational" juries are. The greater the unexplainable portion, the more irrational are juries and vice-versa. In the next section, I give a brief overview of the results of my research.
The Data
To analyze how juries determine awards for damages, I examined data obtained from Jury Verdict Research, Inc. CJVR) on 666 nonfatal, drunk- driving cases occurring in the time period from 1980 through 1990. JVR collects data on a nationwide basis from all 50 states and estimates that it gathers information on approximately 40 percent of all verdicts.4 In addition, to be able to capture socioeconomic factors not available from the JVR data (probable racial composition of the jury. average income, etc.), 1 also used U.S. Census data for 1980 and 1990.
For each case, JVR reports the total award and its components (wage, medical, etc.); the outcome reached (verdict, settlement, etc.); nature of the liability (contributory negligence, punitive damages, mental impairment of the plaintiff and offender, etc.); the age, gender and occupation of the plaintiff; whether the defendant was an individual or institution; a description of the injury; and other data related to the case.
Reported award values from the JVR data ranged from under $1,000 to over $23 million, with a mean of just over $1 million. In order to be able to compare the cases to each other more easily, I used a single, whole-body measurement of impairment, developed for the US Department of Transportation and other Federal Agencies, that incorporates all of the injuries received by the victim and generates an index measurement going from 0 (no impairment) to 1 (complete impairment).5
Research Results
The single most important result of the study is that it is possible to predict with a high degree of accuracy what the jury's award decision will be, based on the observed, objective factors reported in JVR data. Over 75 percent of the variation amongst awards can be explained by these observed factors. In other words, to a surprisingly large degree, juries base their decisions on the actual facts of the case. There appears to be little subjectivity or emotion involved in the decision for damages.
Another key finding is that juries base most of the award on the nature and type of injury. This is highly reassuring since it is the injury itself that gives rise to the lawsuit and serves as the basis for compensatory claims. It is the jury's job to compensate for loss based on the factual evidence presented regarding the injury. This finding was very consistent in my results.
A further key finding is that juries give greater weight to aspects of the injury that they can directly see. That is, victims whose impairments affected their mobility (walking, etc.), senses (e.g., loss of sight), and / or their appearance (visible scarring) received larger awards than victims whose impairments were just as severe on the impairment scale but not so easy to observe, such as cognition (e.g., memory loss, inability to concentrate), pain (chronic neck pain resulting from the accident), and work-related impairments that affect the victim's ability to return to work. This suggests a conservative, skeptical and suspicious approach to injury evaluation.
I also checked to see if the magnitude of the award was consistent with the loss of enjoyment of life or pain and suffering associated with injury. This is an area of jury decision making most frequently criticized by tort reformers. As noted above, included in my data was a comprehensive measure of impairment arising from the injury, with a range from 0 to 1. A rating of 1 essentially equivalent to death.
Using this measure, I calculated the jury's implied value of life. As an example, if a plaintiff has an impairment rated at 0.5 (50% loss) and the jury awards $2 million, then the jury's implied value of life is $4 million, I found that juries calculated awards that implied a value of human life between $2.2 to $4.8 million. This valuation of life is totally consistent with the statistical value of life estimated by academic researchers and used as standards by US Government Agencies and US industry.6,7
Juries take income losses into account in very sophisticated ways, as they should. People in lower-income jobs received smaller awards, as one would expect. Juries are also conscious of the age of the plaintiffs and made the largest awards to plaintiffs who are around 40 years old. This is consistent with accepted economic theory and research: significantly younger plaintiffs are awarded less in general; their careers are less certain and they are likely to have fewer dependents. Significantly older plaintiffs are also awarded less; the present value of their remaining work life is less and they also have fewer dependents.
I also found that juries do not decrease the compensatory component of the award prior to their assessment of contributory negligence by the plaintiff. They seem to be following the judge's instructions to determine the award based on the victim's loss, and then determine if the victim's behavior helped lead to the accident. Further, juries react neutrally to allegations that the defendant was impaired at the time of the accident, even if the defendant is later convicted of driving under the influence. These two findings show a rather incredible ability to mentally separate liability issues from damages measurement.
Juries do appear to reduce compensation for plaintiffs who have been alleged to be impaired at the time of the incident. Apparently, the behavior of the defendants affects issues of liability only, while possible impairment of the plaintiff does affect the award. These results lend further support for the conclusion that juries are conservative in making awards.
Juries appear to compensate for medical costs on a one-for-one basis, suggesting a fairly rational, cost accounting, approach to the verdict. They do, however, increase the compensatory component when punitive damages are assessed. The results also show that when juries are allocating damages, they are not overloading in one area of the award (say, punitive damages) in order to avoid a cap or other limit on damages in another area (compensation).
Finally, juries award more conservatively against individual defendants, perhaps based a perceived lower ability to pay. Since assessing compensation involves estimation, errors on the low side are apparently made regarding awards against individuals.
Conclusion
The overall pattern of results fits that of a system where, in general, rationality and thoughtful process are the standards in many key aspects of decision making. There is nothing in the results that constitutes an urgent call for meddling in the civil jury system.
To the contrary, there is much evidence for trusting juries to be fair and even restrained. Most of the verdict is predictable based on the extent of the injury, medical costs and lost income, indicating rational decision making.
There are several conservative tendencies: Non-observable aspects of the injury are given much less weight. The award is lower if the plaintiff was alleged to be intoxicated or if the defendant is an individual.
Further, juries appear to be scrupulous in not allowing any impact on the award to arise from liability issues involving the defendant's intoxication, the defendant's conviction for drunken driving, or the plaintiff's contributory negligence. Only when egregious defendant behavior warrants punitive damages do compensatory damages tend to become more liberal.
What emerges from this research is a picture of a jury as a deliberative, restrained body, which takes into account relevant information and bases compensation primarily on measures relating to the degree of injury to the body and the ability to earn.
What this study has found is overwhelmingly reassuring about jury behavior. Juries appear to be neither capricious nor liberal, but thoughtful and conservative in general.
Americans are activists. Our democracy was born in rebellion. Sometimes we rush to reform. From Sinclair Lewis' portrayal of the meat industry in the 1880s to current complaints of excess election campaign spending, our institutions are constantly being examined and frequently condemned.
Sometimes the need for reform is legitimate. However, reform should neither precede thoughtful analysis of the problem nor ignore evidence of the system working. Sound evidence and thoughtful analysis are required to ensure that appropriate actions are taken.
The jury system is grounded in a profound trust in the common sense of our fellow citizens. My research shows that placing faith in our fellow citizens may make much more sense than trusting professional politicians to second-guess juries. The system is not only not broken, but it is also revealed to be working remarkably well.
About the Author
Economist Stan V. Smith is president of Corporate Financial Group in Chicago, a nationally recognized economic consulting firm offering consulting services and Litigation support in economics and finance to defense as well as plaintiff counsel nationwide. Dr. Smith received his Ph.D. in Economics from the University of Chicago and has written many articles and co-authored a textbook on Economic/Hedonic Damages. He and his work have been profiled in the Wall Street Journal, the American Bar Association Journal, the National Law Journal, Trial, the publication of the Association of Trial Lawyers of America, and in many law review articles.
Endnotes
- Professor Gary Becker was awarded the Nobel Prize in Economics in 1992. My four dissertation committee members at the University of Chicago were responsible for assuring that I followed sound research practices. The opinions expressed in this paper are my own.
- Viscusi, Kip W., "Pain and Suffering in Product Liability Cases: Systematic Compensation or Capricious Awards?" International Review of Law and Economics. 8, 1988. pp. 203-220
- MacCoun, Robert, "Inside the Black Box. What Empirical Research Tells Us About Decision making by Civil Juries," in Ulan, Robert E., ed., Verdict: Assessing the Civil fury System, Washington. D.C: The Brookings Institution, 1993.
- Cohen, Mark A. and Ted R. Miller. "The Monetary Value of Pain and Suffering due to Criminal Victimization: Evidence from Jury Awards," Working Paper, 1994.
- Miller, Ted R. et al., "The Cost of Highway Crashes," final Report to Federal Highway Administration, The Urban Institute, 1991, and Rossman Shelli B., Ted R. Miller, and John B. Douglass, The Cost of Occupational Musculoskeletal and Severe Occupational Traumatic Injuries. Final Report to NIOSH. Washinglon.D.C.: The Urban Institute, 1991.
- Miller, Ted R-, "The Plausible Range for the Value of Life; Red Herrings Among the Mackerel, "journal of forensic Economics. 3(3), 1990, pp. 17-39.
- Gillette, Clayton P. and Thomas D Hopkins, "Federal Agency Valuations of Human Life," Report to the Administrative Conference of the United Slates. July 7,1988.