Printer Friendly

An exploration of "noneconomic" damages in civil jury awards.


     A. The Challenge of Noneconomic Damages
     B. The Challenge of Assessing the Appropriate Level of
        Noneconomic Damages
     C. Extant Empirical Research on Noneconomic
     A. Data
     B. Analytic Approach
     A. Cook County Data
     B. Bureau of Justice Statistics Data
     C. RAND Jury Verdict Study Data
     A. Insurance Research Council Closed Claim Data
     B. Texas Department of Insurance Data
     C. Missouri Insurance Department Medical
        Malpractice Data


Despite controversy about its role in awarding damages in cases involving punitive damages and medical malpractice, the American civil jury remains a common institution for resolving tort claims and other types of disputes in all fifty states and in federal courts. (1) Not surprisingly, a substantial empirical literature consisting of archival studies, jury simulation experiments, and, in one important instance, records of the actual deliberations of Arizona civil juries, has attempted to learn about how civil juries perform the tasks assigned to them. In this Article, we address a relatively understudied aspect of personal injury awards by civil juries, namely compensatory awards. Specifically, we explore the relationship between so-called "noneconomic" components of compensatory awards and their economic components, a subject that Marshall Shapo among others has discussed as one of the most controversial subjects in tort law. (2) Although previous research, partially summarized infra, has explored the relationships between economic and noneconomic components of jury awards, it has been limited in both scope and empirical evidence. (3) Further exploration of this relationship seems appropriate to the theme of the present conference which seeks to place the American civil jury in its political context. (4)


A. The Challenge of Noneconomic Damages

Tort law provides monetary compensation for losses in personal injury cases that include not only concrete, tangible losses such as medical bills, property loss, and past and future lost income, but also for losses that are clearly tangible but ineffable in monetary terms. These include pain and suffering, loss of society, emotional distress, loss of consortium, disfigurement, loss of child-bearing capacity, loss of parental guidance, and loss of enjoyment of life, as well as other categories of loss. Along with the issue of punitive damages, (5) these latter components of damage awards, often summarized as "noneconomic" awards, or simply--and incorrectly--labeled as "pain and suffering," are among the most contentious issues related to the American civil jury. They play a significant role in attempts to limit awards in medical malpractice cases, (6) are a central target of so-called "tort reform" efforts, (7) and are often based on fallacious claims and distortions. (8) Not surprisingly, they are the subject of frequent legal commentary about the goals of tort law. (9)

A sizeable body of empirical literature bearing on jury damage awards utilizes various research approaches, including systematic interviews with jurors following their verdicts, (10) analyses of actual jury deliberations, (11) simulation experiments with varying factors that might influence the verdict process, (12) and archival studies that draw upon verdict reports to compare the components of actual jury awards across an array of cases. (13)

In this Article, we utilize archival data from multiple sources to attempt to shed more light on the noneconomic components of civil jury awards in personal injury cases. Pursuing this topic by systematically utilizing archival data is important for two reasons. The first reason lies in the nature of noneconomic damages. As many commentators have pointed out, by their very nature noneconomic damages are conceptually a contradiction in terms: they provide monetary compensation for an injury that is intangible in monetary terms. (14) A common claim is that juries do not have the competence to assess these damages because jurors are too often swayed by emotions and in particular do not have the perspective of comparable cases, thus, injecting randomness and unfairness into the tort system. (15)

B. The Challenge of Assessing the Appropriate Level of Noneconomic Damages

Critics claim that the noneconomic portion of awards is often much greater than the actual economic loss, suggesting that emotion rather than reason influences juries. (16) Yet, consider the following case of Lillian Walters, a thirty-two-year-old stay-at-home mother of four minor children. (17)

In December 1979, Mrs. Walters's family physician discovered a lump on her neck, and after conducting some tests referred her to a surgeon. (18) The surgeon advised that Mrs. Walters have a portion of her thyroid gland removed due to its diseased condition. (19) The suggested surgery was relatively low risk and normally would result in a small scar. (20) A day after the surgery, while Mrs. Walters was still in the hospital, her condition deteriorated. Her head ballooned in size, she became blind, and she suffered severe respiratory distress. (21) Shortly after she was moved into intensive care, the pathology department advised the surgeon that a piece of esophagus tissue was attached to the thyroid specimen. (22) It was determined that the area of the surgery was now badly infected, (23) and Mrs. Walters was taken back to surgery where the surgeon reopened the wound, discovered a significant hole in the esophagus, determined that repair was not possible, and sewed the esophagus closed. (24) Initially Mrs. Walters could only be fed by a tube inserted into her stomach through the abdomen; she did regain her vision, and after numerous hospitalizations and surgeries, Mrs. Walters was left with a replacement esophagus fashioned from a portion of colon. (25) Although she was then able to consume food via her mouth, the replacement esophagus did not function like an actual esophagus. Mrs. Walters had great difficulty both swallowing and keeping food in her stomach. (26) Eating was painful, and she could not lie flat because food would come back up through the replacement esophagus. (27) Her condition was embarrassing and distasteful to people around her and made living a normal life impossible. (28) Her life expectancy was more than forty years, but no further medical treatment would improve her situation. (29) Because she did not work outside the home, (30) economic damages consisted entirely of past medical expenses, and those expenses totaled approximately $59,000. (31) What would be an appropriate amount for noneconomic damages in this case? Walters's lawyer asked for $4 million in total damages, (32) and the jury in the case awarded $2 million, (33) which meant that the noneconomic damages were about thirty-three times the economic damages. Was this excessive in this case, where the plaintiff experienced a severe, life-changing event with tangible consequences that she had to endure for the rest of her life? Was the compensation award unreasonable? (34)

The Walters case partly reveals some of the conceptual problems that exist in thinking about noneconomic damages. Although often labeled by critics as merely pain and suffering, should her injury be labeled so simply? In fact, many state legislatures have defined additional elements of damages for which there is no clear dollar value, but the jury or judge translates the injury consequences into monetary terms. A sample of medical malpractice verdicts from the Cook County Jury Verdict Reporter helps to illustrate these issues.

Araujo v. Leong involved an injury during birth causing hypoxic encephalopathy to deep structures in the brain and resulting in severe cerebral palsy and quadriplegia. (35) The jury award was $17,070,000, which consisted of $3 million in past and future medical expenses, $10 million for caretaking expenses, $570,000 for other economic losses, $1 million for loss of a normal life, $2 million for pain and suffering, and $500,000 for disfigurement. (36) Thus in this case the noneconomic component of the award amounted to 8.8% of the total.

Estate of Petre v. Kucich involved a patient who suffered a serious staph infection. (37) The jury award was $814,444, which consisted of $350,000 for loss of a normal life, $200,000 for pain and suffering, $50,000 for disfigurement, $175,000 for emotional distress, and $39,444 for medical expenses. (38) Thus in this case the noneconomic component amounted to approximately 95% of the total.

Estate of Pettway v. Advocate Trinity Hospital involved a two-year-old male child who suffered a seizure and was rushed to the nearest hospital. (39) A decision was made to transfer him to the University of Chicago Hospital, but first a CT scan was needed to rule out bleeding in the brain. (40) However, there were no records of monitoring the child or recording his vital signs. The child died; he was survived by his parents and five siblings. The breakdown of the $3,662,221 verdict was as follows: $7813 for medical expenses, $4408 for funeral expenses, and $3,650,000 for loss of society. (41) The breakdown of the loss of society component was as follows: $1,500,000 each for the mother and father, $250,000 for one sibling, and $100,000 each for four other siblings. (42)

In Hopper v. Lopez, a fifty-seven-year-old male suffered a seizure and was taken to an emergency room. (43) Doctors ordered an MRI to rule out infection as the cause of the seizure. (44) The radiologist who interpreted the MRI failed to report severe sinusitis and an extension of the infection into the left side of the brain. (45) Ten neurologists and an internist relied upon the radiologist's report. (46) The hospital released Mr. Hopper. (47) Two weeks later, however, he had a new set of seizures because the infection had expanded throughout the entire left side of his brain, and five surgeries were required to stabilize him. (48) Mr. Hopper suffered severe, permanent cognitive deficits, which affected every part of his life. (49) The $2,626,000 verdict was composed of $626,000 for past and future medical expenses, $1 million for past and future loss of a normal life, $250,000 for past pain and suffering, and $750,000 to his wife for loss of consortium. (50)

Rodriguez v. Friedman resulted in a $3,270,000 verdict for a brachial plexus birth injury. (51) The child had a shorter, smaller right arm and motion deficits. He underwent three surgeries and several years of physical therapy, and he was recommended for a future surgery. (52) The jury awarded $225,000 for past and future medical expenses, $200,000 for educational expenses, $50,000 for loss of wages, $1,677,000 for past and future disability, $363,350 for disfigurement, and $754,650 for past and future pain and suffering. (53)

The Walters case and the selected other examples draw attention to two important issues. First, the noneconomic aspects of plaintiffs' injuries should not be lumped into a single category called "pain and suffering." Illinois and other state legislatures have delineated a number of categories of damages that, like pain and suffering, cannot be translated directly into a monetary sum. (54) Rather, compensation for these other categories of injury requires human judgment to convert the injury into a monetary sum, typically determined by a jury, although sometimes by a judge or arbitration board. (55)

Second, despite serious or even grievous injury, the plaintiff may have no economic losses or very small economic losses in comparison to noneconomic losses, as the Walters and Araujo cases above help to illustrate. (56) And, of course, the important substantive and methodological consequence of this observation is that using economic loss as the denominator for assessing noneconomic losses can be very misleading because economic loss does not always capture the severity of the injury in terms of the noneconomic consequences of that injury. Still, one would expect that in the aggregate there would be a relationship between economic and noneconomic injuries and the consequent total amount of the damage award. The core question we examine next is the nature of the relationship between economic damages and noneconomic compensatory damages as determined by juries.

C. Extant Empirical Research on Noneconomic Damages

The second reason for our focus is that the empirical literature bearing on what juries actually do in regard to these noneconomic claims is surprisingly sparse.

Jeffrey O'Connell and Rita Simon looked at payments for pain and suffering using a sample of claims that a single insurance company paid to Illinois residents in 1966. (57) They limited their study to claims involving a payment of at least $100, including property loss, economic damages, and noneconomic damages. (58) Their data were derived from a combination of insurance records and interviews with 391 claimants. (59) The authors computed the ratio of payment to loss--the "recovery ratio"--for cases that were litigated (only 17 such cases were in the sample), unlitigated cases in which an attorney represented the claimant (77 cases), and cases with no representation (297 cases); (60) the respective recovery ratios for the three groups were 5.3:1, 2.1:1, and 1.5:1. (61)

In an early study of jury verdicts, Mark Peterson examined almost 9000 cases from the 1960s and 1970s compiled in the Cook County Jury Verdict Reporter. (62) Among his findings, verdicts involving high medical expenses and lost income were approximately 4.5 times larger than verdicts for plaintiffs having lesser injuries. (63) Plaintiffs with medical malpractice, product liability, and work injury claims obtained two to four times more than plaintiffs with other types of personal injury claims. (64) However, Peterson was only able to look at total compensatory damages because the data he employed did not separate out various types of compensatory damages. (65)

Ostrom and his coauthors examined data from the National Center for State Courts's study of forty-five urban trial courts during 1992. (66) Those researchers found a large discrepancy between the mean and median awards--with the mean being greater than the median (67)--indicating that very large awards had skewed the distribution. Medical malpractice cases tended to have much higher awards, on average, than other types of cases. (68) Again, the researchers did not separate economic and noneconomic components of the damages.

Danzon's and Lillard's study of a sample of medical malpractice cases from liability insurers' files closed in 1974 and 1976 found that approximately 7% of claims went to trial and plaintiffs prevailed 28% of the time. (69) Comparing the awards with estimates of economic losses, injury severity, and the plaintiff's age, Danzon and Lillard found that total jury awards for compensatory damages were related to the magnitude of the plaintiff's losses. (70) Those authors assumed that the difference between the insurers' measure of economic loss and the jury award constituted the jury's award for noneconomic damages. (71)

Bovbjerg and his coauthors analyzed a sample of 898 personal injury cases that went to a jury. (72) The median award in 1987 was $82,000, but the mean award was $490,000. (73) Those authors also assumed that the difference between economic loss and total award constituted the compensation for noneconomic damages. (74) Their study also coded the seriousness of the injury according to the National Association of Insurance Commissioner's (NAIC) nine-point scale of injury severity. (75) Awards increased with the severity of injury, except when the outcome was death in which case the award was typically lower. (76) Severity of physical injury accounted for about two-fifths of the variation and other factors accounted for one-fifth. (77) The authors speculated that jury unreliability may explain most of the remaining variability. (78)

Daniels and Martin compared medical malpractice and product liability awards in a large sample of cases reported in verdict reporters from venues around the United States. (79) They also found that awards were related to severity of injury. (80) They did not look explicitly at the awards for noneconomic damages because virtually none of the verdict reporters that constituted their sources reported that information. (81)

Viscusi compared payments in a sample of product liability cases, most of which were settled rather than tried. (82) He concluded that payments were related to severity of injury. (83) However, for several reasons, Viscusi's analysis tells us little about jury behavior. First, he did not separate out results for cases in which juries awarded damages to the plaintiff (only 1.5% of the filed claims in his dataset--roughly 150 cases--resulted in a court verdict for the plaintiff). (84) Second, it appears that his analysis focused on payments, which means that even for the cases where there was a plaintiff's verdict, the actual payment may have reflected a remittitur or a post-verdict settlement rather than the amount set by the jury. (85) Finally, Viscusi had to assume that the payment for pain and suffering was the difference between the amount paid and the insurer's estimate of financial loss. (86)

Taragin and his coauthors analyzed a sample of cases taken from the New Jersey Medical Inter-Insurance Exchange. (87) Those investigators were mostly interested in estimates of defendant responsibility, (88) but they did find a relationship between rating of injury severity and amount of total payment. (89) However, their research did report separate elements of the payment data. (90)

Sloan and his coauthors found that in a sample of medical malpractice cases levels of injury and economic losses varied substantially from patient to patient, even among those with roughly comparable injuries. (91) While such variability should surprise no one who thinks about such factors as age and economic differences between plaintiffs, this explanation frequently has been ignored.

Finley examined a sample of California medical malpractice jury verdicts and the potential effects of California's MICRA cap on pain and suffering on plaintiffs who were children, women, elderly persons, and members of minority groups. (92) Finley argued that these were plaintiffs most likely to have relatively low economic losses but major claims for noneconomic damages. (93) Her conclusion was that caps unfairly disadvantaged these types of plaintiffs. (94)

Vidmar, Gross, and Rose obtained a sample of jury verdicts in medical malpractice cases from jurisdictions within three different states: New York, Florida, and California. (95) Those authors classified the cases according to injury seriousness. Consistent with previous research, the total awards were positively related to the seriousness of the physical injury suffered by the plaintiff but tended to drop in cases involving death of the patient. (96) However, similar to previous studies, those authors found that verdict reporters seldom listed the specific elements of the general damage award, instead often lumped all noneconomic awards as pain and suffering. (97) Nevertheless, there were exceptions, namely reporting awards for such losses as loss of companionship, loss of consortium, emotional distress, disfigurement, mental anguish, loss of enjoyment of life, and human damages. (98)

One of the few studies that looked specifically at noneconomic awards was Vidmar and MacKillop's study of a sample of medical malpractice cases from Cook and DuPage counties, in Illinois, as well as two downstate counties covering the years 2001 through 2004. (99) Their article addressed the potential effect of a cap on pain and suffering during that time period. (100) The findings clearly contradicted the claims that pain and suffering constituted 90% of all malpractice verdicts. (101) For the year of 2001, there were thirty medical malpractice awards in Cook and DuPage counties. (102) While eight cases involved pain and suffering awards that equaled or exceeded $1 million, (103) in at least five other cases there was no pain and suffering award. (104) And the pain and suffering component exceeded the economic losses in only four cases. (105) One calculation suggested that, on average, pain and suffering constituted only 15% of the award. (106)

This summary of the extant empirical literature on compensatory damage awards indicates that empirical evidence is sparse, especially when it comes to the noneconomic components of those awards. As Vidmar observed, data in verdict reports often have substantial weaknesses. (107) Much of the difficulty lies in the absence of data, especially information on the precise nature of the injury and breakdowns of awards into economic and noneconomic components. (108) Moreover, even when the data sources do separate economic from noneconomic components, they rarely delineate the particular elements of noneconomic awards. (109) Nevertheless, with appropriate qualifications, archival data can provide an important starting point to our understanding of what juries actually do.

In this Article, as described immediately below, we draw upon various sources to estimate the relative percentages of jury damage awards across time and across case types. Our goal is to increase discussion about what civil juries do in awarding damages. Hopefully, the findings will provoke discussion about the causes of these outcomes and their fairness--or lack of it--and lead to insights about the role of civil juries in the American democratic process.

II. Data Sources and Analytic Approach

A. Data

As noted in the Introduction, scholars have conducted fairly extensive analyses of punitive damage awards. (110) This has been facilitated by the fact that when there is a request for punitive damages, the jury is asked to decide whether such damages should be awarded and, if so, the specific award. (111) As suggested by our brief review in the previous section, the challenge in looking at noneconomic damage awards is that in many courts juries are asked to return a general verdict in which only a single figure is given covering all compensatory damages. (112) Specific figures for categories of compensatory damages exist only when a jury has been presented with a special verdict form with those categories specified. (113) There are three notable situations in which a special verdict is used. The first is if there is a cap on one or more categories of compensatory damages, such as in medical malpractice cases in some states. (114) Normally, juries are not explicitly told of such caps, and the judge will reduce any amount above the cap to no more than the cap. (115) The second situation arises when there is a local practice of using a special verdict form for damages in personal injury cases. (116) The final situation occurs when one side in a case specifically asks that a special verdict form be used--perhaps when the defense is concerned that an award will be excessive--and believes that having a breakdown into categories will facilitate a request for a remittitur. (117)

We identified three sources of data compiled directly from jury verdicts which contain useful information on both economic and noneconomic damages. (118) The first source is a set of original data compiled by the authors using the Cook County Jury Verdict Reporter. We obtained copies of verdict reports for auto accident, medical and dental malpractice, and premises liability cases for the years 2005 and 2010. (119) From these reports we coded the amounts for each detailed category of damages listed in the report. We also coded the gender of the plaintiff, the age of the plaintiff, and the severity of the claimed injury using the NAIC injury scale. (120) Coding was conducted by a staff assistant and checked by one of the authors. In cases where multiple plaintiffs suffered personal injuries, we treated each plaintiff as a separate case for purposes of analysis, which gave us a total of 262 observations. (121) The amounts from the 2005 verdicts were adjusted to 2010 dollars.

Not all of the case reports had full breakdowns of damages; for some of those cases we inferred the breakdown between economic and noneconomic damages in one of two ways. For cases that showed specific breakdowns for categories of economic damages or categories of noneconomic damages, but not for both, and for which the total of the breakdown reported was less than the overall verdict, we assumed that the difference between the total and the verdict was the other type of damages. For example, if the total verdict was $20,000 and the report showed that $8000 was for past medical expenses, and $3000 was for lost income, we assumed that $9000 was for noneconomic damages. For cases that provided no breakdown at all but presented the amounts the plaintiff claimed for medical expenses and/or lost income, and the sum of those amounts was less than the verdict, we assumed that the difference between the verdict and the claimed economic damages was the noneconomic damages component. Thus, if the report showed the verdict as $20,000 and the plaintiff claimed $8500 in economic damages, we assumed that the remaining $11,500 awarded was for noneconomic damages. This provided us with 205 cases for analysis. Finally, we relied on the gross jury awards before any adjustments for comparative negligence.

The second source is data on civil jury and bench trials collected by the Bureau of Justice Statistics (BJS) and the National Center for State Courts (NCSC) as a part of the Civil Justice Survey of State Courts (CJSSC) involving a sample of counties around the United States for the fiscal year 2005. (122) BJS and NCSC collected data from samples of the seventy-five largest counties in 1992, 1996, 2001, and 2005; (123) in 2005, BJS and NCSC extended the study to include a sample of nonurban counties. (124) Only the 2005 dataset includes information on both economic and noneconomic damages, and hence, we employ only that dataset in our analysis (henceforth "BJS dataset").

The third source is from the RAND Institute of Civil Justice jury studies project. (125) RAND collected data from local jury verdict reporters in a series of waves. (126) RAND archived the earlier sets of data covering the period 1960-1984 with the Inter-university Consortium for Political and Social Research. (127) Unfortunately, those data do not include separate information on economic and noneconomic damages. In a later collection, covering the period 1995-1999, RAND did ask its coders to capture separate information on economic and noneconomic damages when that information was available. (128) The data come from selected counties in six states: California (forty-six counties), Illinois (Cook County only), Texas (Harris County only), Missouri (St. Louis City and County, Jefferson County, and St. Charles County), New York (fifty-three counties), and Washington (King County only). (129) RAND has not archived the later dataset, and that information has been largely unanalyzed. (130) However, RAND generously provided us access to this dataset. Included are 2301 personal injury cases from selected counties in six different states resulting in plaintiff's verdicts that include information on both economic and noneconomic damages. (131)

In addition to the three sources compiled directly from reports of jury verdicts, we have also identified three other sets of data derived from insurance company files dealing with tried cases that provide some information on noneconomic damages. Because each of these datasets has a major limitation, however, we view our analyses of these latter sources as supplemental and thus, only report the results in the Appendix, accompanied by our analyses.

B. Analytic Approach

We modeled the core analysis that follows after the recent analyses of the relationship between punitive and compensatory damages reported by Eisenberg and his colleagues. (132)

Specifically, consistent with Eisenberg, we looked at the relationship between economic and noneconomic damages graphically by fitting a simple linear regression line. In order to deal with the extreme ranges involved and the relative infrequency of very large amounts, we transformed the amounts of damages to logarithms. (133) We also looked at the ratio of noneconomic to economic damages conditional on the amount of economic damages; many of our tables are modeled after Table 1, which appeared in one of Eisenberg's recent articles (134) and which uses data from BJS and NCSC studies for 1992, 1996, and 2001. (135)


A. Cook County Data

One feature of the Cook County data is that most reports provide detailed breakdowns of the damage awards, both for economic damages and noneconomic damages. Table 2 shows the breakdown; we have limited the information shown in Table 2 to the 200 cases where the breakdown was complete and the sum of the various categories of damages equaled the amount shown for the total verdict. The table shows both the breakdown for all of the cases in the sample and for the three separate categories of auto accidents, medical and dental malpractice (henceforth "medical malpractice"), and premises liability cases.

The table shows six types of expenses that fall under economic damages, including dental expenses, education expenses, funeral expenses, medical expenses, lost income/time/wages, and miscellaneous. It also shows eight types of expenses that fall under noneconomic damages, including disability, disfigurement, emotional distress, loss of consortium, lost normal life, loss of society, pain and suffering, and miscellaneous. Overall, economic damages made up 43% of overall awards, and noneconomic damages constituted 57% of total awards. However, the proportion of economic and noneconomic damages differed across type of cases. For auto cases, the awards were split evenly between economic and noneconomic damages. In comparison, the ratios of economic to noneconomic damages were 1:3 for medical cases and 2:3 for premises liability cases.

As one would expect, medical expenses seem to dominate economic damages in that the majority of cases reported medical expenses. About 96% of auto cases (124 out of 129), 72% of medical malpractice cases (39 out of 54), and 82% of premises liability cases (14 out of 17) listed

medical expenses. In terms of dollar values, medical expenses accounted for 44% of total damage awards for auto cases, 21% for medical malpractice cases, and 26% for premises liability cases. The second most frequent type of economic damages was lost income, both past and future: 33% of auto cases, 22% of medical malpractice cases, and 53% of premises liability cases reported lost income. In terms of dollar values, lost income tended to be a small part of the overall award, making up on average 5% of total damage awards for auto cases, 2% for medical malpractice cases, and 12% for premises liability cases. The remaining four types of economic damages were minimal for all cases.

Pain and suffering dominated noneconomic damages, and the majority of cases reported such damages. About 85% of auto cases (110 out of 129), 75% of medical malpractice cases (42 out of 54), and 88% of premises liability cases (15 out of 17) listed pain and suffering damages. In terms of dollar values, pain and suffering accounted for 32% of total damage awards for auto cases, 31% for medical malpractice cases, and 38% for premises liability cases. The other category of noneconomic damages that was fairly common was loss of normal life: 46% of auto cases, 50% of medical malpractice cases, and 53% of premises liability cases include awards for loss of normal life as noneconomic damages. In terms of dollar values, lost normal life accounted for 11% of total damage awards for auto cases, 14% for medical malpractice cases, and 13% for premises liability cases. In addition, a substantial number of cases listed disfigurement and disability, and the pattern was consistent across types of cases. Loss of society seems to only have mattered for medical malpractice. For example, 24% of those cases listed loss of society as noneconomic damages, and the award amount of loss of society made up 15% of total damage awards. In comparison, loss of society was minimal for both auto and premises liability cases.

Turning now to the relationship between economic and noneconomic damages in the Cook County dataset, we looked at all cases together and then split the cases into auto accidents, medical malpractice, and premises liability. Figure 1 shows the plots for the 206 cases for which we had information on both economic and noneconomic damages, (136) and both were nonzero; the figure shows the plot for all cases and for the three separate categories. Each plot shows two lines. The broken line is a least squares linear regression line. We fitted the solid line using Stata's locally weighted scatterplot smoothing procedure (LOWESS, also known as LOESS). Essentially, LOWESS fits a series of short lines using overlapping subsets of the data, which can illustrate nonlinearities without requiring the specification of a particular functional form. If the LOWESS and linear regression lines are very close, it is a good indicator that the relationship is linear. Figure 1 shows a strong linear relationship for all of the cases taken together and for auto cases. The fits for the medical malpractice and premises liability cases appear to deviate from linearity, with the medical malpractice cases not following a clear pattern. This may reflect the relatively small number of cases, particularly the small number of premises liability cases (sixteen). Table 3 summarizes the regressions represented by the broken lines in Figure 1. Not surprisingly, the regression fit for medical malpractice was very weak; the fit for premises liability was better, but not as good as for the auto accident cases.

Because we have an indicator of the severity of the injury in the form of the NAIC codes for the Illinois data, we can assess whether using the additional information helps to predict the noneconomic damage award. We also have the gender of the plaintiff and, for most of the observations, the plaintiff's age. Preliminary analysis showed that there were no statistically significant differences between auto accident and premises liability cases, and hence we collapsed those two categories for purposes of the extended analysis.

Table 4 shows a set of four models that employ various combinations of variables. Model 1 in Table 4 includes the predictors including the logarithm of the economic damages, the injury severity measure, the interaction of economic damages and injury severity (that is, an indicator of whether the influence of one of these variables depends on the other), a dummy variable for medical malpractice cases, a dummy variable for gender, the plaintiff's age, and the square of the plaintiff's age. (137) The fit of the model is quite good; it accounts for 69% of the variation in the logarithm of noneconomic damages, a substantial increase over the 53% explained by economic damages alone. (138) All of the predictors were statistically significant except for age and gender.

The negative interaction term indicates that as economic damages increased, the impact of severity decreased (or alternatively, as severity increased, the impact of economic damages decreased). The coefficient for medical malpractice indicates that, with other factors held constant, noneconomic damage awards were higher in medical malpractice cases. However, as indicated in Model 2, which adds an interaction between medical malpractice and economic damages, noneconomic damages increased with economic damages less slowly in medical malpractice cases; in fact, combining the economic damages coefficient and the interaction with medical malpractice for Model 2, the effect of economic damages in medical malpractice cases was about half of what it was in auto and premises liability cases. In this model, the interaction between injury severity and economic damages is no longer statistically significant, and the two demographic variables remain nonsignificant.

Models 3 and 4 modify how severity is handled by treating death cases differently. In these models, a modified injury severity index codes death as zero with a separate dummy variable to indicate death; in addition, the model includes an interaction term between the death dummy variable and economic damages. With these changes, the modified injury severity index in Model 3 does not achieve statistical significance while the death indicator is strongly significant; neither of the interaction terms involving the injury and economic damages is statistically significant, nor are the demographic variables.

Model 4 drops gender and age, which had shown no evidence of having an influence on noneconomic damages; this adds a small number of additional cases to the analysis. This model explains 73% of the variation in noneconomic damages, and all predictors except for the interaction between economic damages and the modified injury severity index achieve statistical significance at the .05 (one-tailed) level or better.

Table 5 shows the ratio of noneconomic to economic damages broken down by level of economic damages and type of case. Unlike in Table 1, which showed a similar type of breakdown for the ratio of punitive damages to compensatory damages, we do not observe a consistently declining ratio as the amount of economic damages increased. As Table 5 shows, this was true for auto accident cases but not for medical malpractice cases and premises liability cases, both of which did show a declining ratio. Exactly why this was not true for auto cases is not clear, but we speculate that it has something to do with the auto accident cases that actually get to trial in Cook County. Specifically, it may be that the larger auto cases that were tried involved particular issues with regard to noneconomic damages, while the smaller cases may have involved issues related to liability or preexisting conditions. As we will show in later sections of this Article, we saw somewhat similar patterns with regard to auto accidents in some of the other datasets we examined.
COPYRIGHT 2014 College of William and Mary, Marshall Wythe School of Law
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2014 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:Introduction through III. Results A. Cook County Data, p.971- 999; The Civil Jury as a Political Institution
Author:Kritzer, Herbert M.; Liu, Guangya; Vidmar, Neil
Publication:William and Mary Law Review
Date:Mar 1, 2014
Previous Article:What's it worth? Jury damage awards as community judgments.
Next Article:An exploration of "noneconomic" damages in civil jury awards.

Terms of use | Privacy policy | Copyright © 2022 Farlex, Inc. | Feedback | For webmasters |