Empirical Tort Law (and Theory) – An Essay in Honor of Deborah Hensler

2.1 Approach

In February 2018, we conducted an initial screener survey administered to a nationally representative sample of 17,161 adult Americans to identify individuals who suffered significant injuries or illnesses in 2017 that met certain criteria. The respondents were members of a probability-based Internet panel developed and managed by what is now known as Ipsos Public Affairs that is widely used to conduct nationally-representative surveys. The significant conditions of interest to our study were defined by criteria such as taking four or more days off work or school, being unable to perform regular activities at home for four or more days, visiting a healthcare provider three or more times, spending at least one night in a hospital or other medical facility, or making at least one visit to the emergency department. A total of 2,967 panelists reported that they had at least one significant medical condition with onset or discovery in calendar year 2017, and provided some descriptive information about a total of 3,343 separate individual-specific conditions.

The first follow-up survey in May 2018 collected detailed information about each reported significant medical condition. The information collected included the manner of injury or illness onset; the extent and severity of the condition; the types of treatment received; the health-care related expenses incurred, paid, or still owed; work and school time lost; sources of money used to address the financial impacts of the condition; the panelist’s perception of responsibility, reasons for pursuing or not pursuing liability-based claims; the types and amounts of liability-based compensation received; satisfaction with processes and outcomes; and attitudes about the civil justice system. Because efforts to seek liability-based compensation can take months or years to develop, a second follow-up survey was distributed in August 2020. This one focused solely on questions related to responsibility, claiming, disputes, lawyer contact, and litigation in order to update what we already knew about these medical conditions from the 2018 survey.

Our analysis of the survey data covered the incidence rates of such conditions, the progression through the dispute pyramid from injury or illness onset to litigation, and the factors influencing each step – including blaming others, seeking compensation, and hiring attorneys. We also explored rates of use of the workers’ compensation system, health insurance, and paid leave, as well as factors affecting use of these support systems.

2.2 To What Extent Do Significant Injuries and Illnesses Result in Litigation and How Do Lawyers Affect That Process?

One of our primary goals was to determine to what extent significant injuries and illnesses give rise to litigation. To do so, we use the concept of the dispute pyramid, a way researchers have conceptualized the stages by which tortious harms can lead to litigation. In our version of the pyramid (see Figure 1), someone incurring a significant injury or illness may initially believe that another party was fully or partially responsible for the medical condition. Continuing to move up through the levels of the pyramid, the agrreved individual could then decide to actively seek tort compensation, contact a lawyer for advice, hire counsel, and ultimately file a lawsuit. We report the relative volume of each layer of the dispute pyramid below:

Figure 1:

Dispute pyramid for significant injuries and illnesses. Values reflect follow-up survey weights. Unweighted Ns excluding non-responses: responsibility = 2,636; compensation seeking = 589; attorney contact = 126; attorney hire = 74; lawsuit filing = 121. Some respondents reporting the filing of a lawsuit may have done so without hiring counsel.

We find a significant drop-off in every step of the dispute pyramid process – from believing that another party was responsible, to seeking compensation from the responsible party, to filing a lawsuit.^[3] For the entire universe of significant illnesses and injuries that our respondents faced, 26 percent resulted in the belief that another party was responsible, 7 percent had such belief and sought compensation from that entity, and only 2.4 percent filed a lawsuit.

We also wanted to understand the role of lawyers. We found that 4.1 % of respondents had contact with an attorney with 3.1 % hiring one.

We found that the first two steps in the pyramid are where the most falloff occurs. Respondents believed that someone else might have been responsible in about a quarter of injuries and illnesses (26.1 %). Even with a belief that someone else might have been responsible, most respondents did not seek compensation from the party (or parties. Less than three out of 10 of those injuries and illnesses in which another was viewed as responsible 27.0 %) resulted in the respondent seeking compensation from responsible parties.

There is somewhat less drop-off once compensation is sought. Of the respondents who sought compensation from a responsible party, 57.7 % attempted to get help from an attorney. About three-quarters (77.2 %) of people who contacted an attorney hired counsel and about three-quarters 76.0 %) of people who hired an attorney filed a lawsuit.

2.3 What Leads People to Blame Others, Seek Compensation, and Use the Civil Justice System to Seek Compensation?

A necessary precondition to filing most lawsuits (and the first step up from the harm-causing incident itself in the dispute pyramid) is blaming, i.e., believing that another party was at least partially responsible for causing the injury or illness or making it worse.^[4] When the injuries and illnesses were medical treatment-related, work-related, product-related, or motor vehicle-related, individuals were more likely to believe that another party was responsible.

The consequences of the injury and illness also were correlated with the blaming rate. Conditions that resulted in payment of medical expenses, use of health insurance, work-related losses, changes in general health, impacts on household work, and impacts on quality of life had much higher blaming rates than conditions that did not. We also observed preliminary evidence supporting our hypothesis that lack of health insurance may make it more likely that a person with a significant health condition takes that first step on the pyramid – blaming others. While the overall blaming rate was 26.1 %, the blaming rate for those who used health insurance to pay medical expenses (23.7 %) was lower than the average. The rate for those who did not use health insurance to pay for medical expenses (27.3 %) was higher than average.

The next step to litigation is seeking compensation from responsible parties (the second step in the dispute pyramid). Here, the context of the injury or illness was important. As we observed with blaming, medical treatment-related, work-related, product-related, and motor vehicle-related injuries and illnesses had much higher rates of compensation seeking compared to the overall average. While severity of injury or illness was not associated with blaming, severity was associated with seeking compensation: between 26 % and 40 % of individuals who described their injuries or illnesses as not serious at all, not too serious, somewhat serious, and very serious sought compensation, whereas 93.4 % of respondents who suffered an extremely serious injury or illness sought compensation.

We asked those individuals who blamed others but did not seek compensation why they didn’t seek compensation. The most common reasons cited reflect respondents’ belief that seeking compensation did not seem to be the right thing to do or was not necessary (e.g., “received all of the monetary help I needed from other sources”, “just an accident and no one was really at fault,” “didn’t want to start trouble”). Other common reasons reflect a sort of informal cost-benefit analysis: Respondents thought that seeking compensation might not be worth it if the relative costs of the process are high or relative benefits are low or because other sources of compensation are available (e.g., “didn’t think anyone could prove how I became ill or injured,” “wasn’t worth the effort I’d have to put in,” “the whole process would be too expensive”).

We also examined the rate at which individuals contacted an attorney for those who decided to seek compensation (another step on the compensation pyramid). We found that 43.3 % of respondents who sought compensation from a responsible party contacted an attorney for assistance with this effort. We found no correlation between income and contacting or hiring an attorney for assistance seeking compensation. We asked individuals who contacted an attorney how they found their attorneys. The most common ways people found an attorney were through recommendations from family, a friend or coworker, and through TV, internet, the radio, or newspaper. As compared to ICJ-Hensler respondents, our survey respondents were more likely to find attorneys through legal advertising and much less likely to find attorneys through personal recommendations.

Interestingly, the injured survey respondents we polled were much more likely to be turned away by the attorneys they contacted when compared to the ICJ-Hensler respondents of 1988–89 (53 % turned away vs. 22 % turned away). When we asked the turned-away respondents why they were turned away, we observed an informal cost-benefit analysis, this time on the part of the attorneys. Attorneys were unlikely to take a case if they believed litigation costs outweighed benefits to the injured or ill party and/or the potential attorney (e.g., “too difficult or expensive to prove the seriousness of the injury or illness,” “too difficult or expensive to prove that the injuries or illnesses were related to what happened,” “too difficult or expensive to prove that some other person or some other company was legally responsible for what happened,” “couldn’t get enough money to make it worthwhile,” “too expensive to take this kind of case to court”).

Just under three-quarters (73.5 %) of people who contacted an attorney hired an attorney^[5] and just over three-quarters (77.6 %) of people who hired an attorney filed a lawsuit.

2.4 Do Individuals Suffering from Illnesses and Injuries Use the Workers’ Compensation System to Cover Losses?

We found that workers’ compensation was not often utilized for injuries and illnesses that occurred on the job: only 29 % of those who were injured or became ill on the job filed worker’s compensation claims. When we asked people who were injured or became ill on the job why they didn’t file a workers’ compensation claim, the most common reason people gave was that “it wasn’t anyone’s fault”. This is surprising, because workers’ compensation is explicitly a no-fault system that does not require the worker to show fault as a precondition for compensation.

Frankly, we are somewhat puzzled by our findings. One possibility is that workers’ wrongly believe that they must show fault in order to seek compensation. This suggests that education might increase use of workers’ compensation.

A second possible explanation for the reluctance of workers to file a workers’ compensation claim is that there are societal norms against filing claims in the absence of fault. Finally, a third possibility is that our survey is picking up illnesses like the common cold and flu that employees believe they contracted at work. These illnesses, even if contracted at work would not traditionally be thought of as compensable by the workers’ compensation system.

2.5 To What Extent Do Individuals Who Suffer from Illnesses and Injuries Use Other Forms of Compensation (Insurance, Paid Leave)?

We observed wide usage of medical insurance to cover medical expenses. 72.9 % of respondents used some type of health insurance to pay for medical expenses. We did not observe wide variation in use of health insurance by income level. Comparing our survey results with ICJ-Hensler study results, we observed that health insurance is slightly more utilized to cover injury-related medical expenses today as compared to 30 years ago (74 % vs. 67 %, respectively).

We also observed modest use of paid sick leave and vacation days to cover work losses. 42.3 % of the injuries and illnesses captured in our survey resulted in a work loss – for example, missing days of work or stopping work completely.^[6] Unsuprisingly, paid sick leave and paid vacation days were by far the most common ways people covered work losses, with 47.1 % of work loss covered by paid sick leave and 21.1 % of work losses covered by paid vacation days.

3.1 Understanding the Claiming and Compensation Process and Reforms That Affect Them

First, we observe a significant incidence of serious illness and injury that befalls our nationally representative respondents. Our survey found that almost one in five Americans suffers such a hardship. On the one hand, this rate of injury or illness is unsurprising – sickness and injury are an oft noted part of the human condition. On the other, the fact that one in five Americans suffer from a significant illness or injury in any given year has real policy implications. Policies that are designed to improve societal welfare either in the civil justice system or elsewhere must account for this significant rate of illness and injury. If the rate of injury and illness were appreciably smaller, the size of our health insurance, workers’ compensation system and our civil justice system could be smaller, and other methods to compensate those individuals might be viable. The frequency and magnitude of injuries and illnesses highlights the important role that compensatory systems play.

Second, in the vast majority of significant injuries and illnesses (74.2 %) victims do not believe that other parties were responsible, much less claim against them. As a result, the civil justice system is largely irrelevant as a means of compensation for these losses.^[7] The tiny percentage of injuries and illnesses for which tort suits are filed highlights the vital role of institutions outside the civil justice system -- the health insurance system to cover medical costs and other programs (sick leave, workers’ compensation, social security, and unemployment insurance) to cover losses when significant illness or injury occurs. Of course, seeking compensation from responsible parties occurs with the implicit threat of litigation so tort judgements surely influence the process of seeking that compensation. But since victims usually believe that no one was responsible for their harm, even the indirect effects of tort play a small role in compensating individuals for their losses. That is to say, the civil justice system makes a terrible social welfare insurance system simply because it compensates such a tiny fraction of the universe of individuals that are injured or ill.^[8]

3.2 Debates About the Role of Tort Law

What do our findings contribute to debates about tort law? How should tort law affect regulation and what is the best way to conceptualize tort either descriptively or normatively? We address these questions in this section.

3.3 Effect of Health Insurance Changes on Compensation and the Claiming Process

We anticipated that the passage of the Affordable Care Act and the wider debate about the importance of health insurance would have significantly increased health insurance as a mechanism for compensating victims as compared to the ICJ-Hensler study 30 years ago in 1991. Surprisingly, the difference in the use of health insurance is relatively small. In 1991, the ICJ-Hensler study found that 67 % of respondents used health insurance for medical expenses related to injuries.^[28] In comparison, when we looked at medical conditions in our data that were defined as injuries we found 72.9 %, an increase of 5.9 %. So, while there was an increase, it was more modest than we expected. The finding also highlights the fact that 27.1 % of our respondents with a significant injuries did not use health insurance for their medical expenses. Recent estimates of the rate of uninsured in the United States are approximately 8 %.^[29] We are uncertain as to the reasons for this relatively low rate of utilization of insurance and further research on this topic would be useful.

3.4 Implications on the Selection of Cases for Litigation and the Common-Law

The selection of cases for litigation has implications for the development of the common law. If only certain types of cases (or cases with particular litigants) are litigated to finality, the court of final appeal will only receive briefing on those cases and may have limited information about the effect of their rulings on the larger universe of cases that are not appealed to their court. Given the financial and time cost of extensive litigation, some have suggested that deep-pocketed repeat player litigants influence the common-law to their advantage in this way.

But this outsized influence over the common law could also be true for specific categories of cases. If motor vehicle injury cases or injuries occurring in the home, for example, were often litigated to finality, tort law might be shaped by the particular problems raised therein.

And while many have noted the greater influence of cases litigated to final resolution, it is true at every level of the dispute pyramid. If only certain types of injuries and illness are likely to result in a victim that identifies another party as responsible, only those types of risks will be subject to litigation and shape the common law. If only certain types of injuries and illnesses result in claims against other parties, only those types of risks will be able to shape the common law. Throughout the litigation process, there is a truncation of the universe of risks that shape the common law. This winnowing process results in a tiny fraction of potential risks being presented to the court of finality.^[30] While there is considerable theorizing about the way this common law process functions, there is little empirical examination of this dynamic.

To the extent the common law shapes public and professional (lawyer) values about when a case is worth pursuing, there may be a self-reinforcing dynamic between the cases that shape the common law and the public/professional opinion. That is, if the public only believes that certain types of cases are worth pursuing (e.g. serious cases that involve some moral blameworthiness), those are the only cases that will be seen by the courts and the only cases that are reported in the media or that have appellate opinions written. Similarly, if lawyers only believe that certain categories of cases are worth bringing, they will also shape the cases the courts see.^[31] In this way, norms about what makes a worthy case can be perpetuated and create a feedback loop.

And there is no particular reason to believe that this dynamic – perpetuation of certain kinds of litigation and truncation of other cases courts see – will yield normatively desirable legal rules.^[32] For all the genius of the common law, even the most conscientious judge can’t know what she never sees.^[33] There may be evolution as litigants and creative lawyers push new variations, but there may also be considerable undesirable path-dependence in the evolution of the common-law.^[34]

While some legal theorists have noted the potential impact on the selection of cases for litigation on the common law, none have attempted to couple their observations with actual data about the process. So, what insight do our findings yield on this shaping process and how it might affect the development of the common law? Below, we proceed step-by-step through the dispute pyramid truncation process and discuss its implications.

As noted above, a precondition for litigation is usually believing that another party is responsible. This occurs in 24.4 % of cases in our sample. Injuries and illness that were medical treatment-related, work-related, product-related, and motor vehicle-related were more likely to result in the victim believing that another party was responsible.^[35] Similarly, the consequences of the injury or illness were also important. Conditions that resulted in payment of medical expenses, use of health insurance, work-related losses, changes in general health, impacts on household work, and impacts on quality of life had much higher blaming rates (though self-described severity did not make a difference). Hence injuries and illnesses of those types and from those contexts are more likely to shape the development of the common law.

At this level of the dispute pyramid the selection effects seem mostly intuitive – cases with more obvious responsible parties and effects are more likely to lead to claiming and thus the possibility of litigation and effect on the common law. As noted above, the obviousness of the responsible parties is a self-reinforcing phenomenon. Victims of motor-vehicle crashes are more likely to bring lawsuits because of the well-developed infrastructure for bringing those lawsuits which, in turn, results in a well-developed body of law and social expectations around those harms. In contrast, many harms that occur within the household lack similar infrastructures and narratives of compensation.

This pattern of self-reinforcement is inconsistent with theories of the common law that suggest that new claims are most likely to be brought in the uncertain areas of the law such that the common law gradually clarifies more and more uncertain legal doctrine. In the real world we observe claims progress because there are existing avenues for those claims that both victims and lawyers are familiar with. Legal doctrine is at least partially shaped by the crucible of the selection of these cases for litigation.

The next step of truncation on the path to litigation in the dispute pyramid is not just believing another party is responsible but seeking compensation from that party. This occurs in only 6.3 % of the incidents we observe in our sample, and 32.1 % of incidents in which another party was believed responsible.

Here we found that the context and severity of the injury or illness was important. While between 26 % and 40 % of individuals who described their harms as not serious at all, not too serious, somewhat serious, and very serious sought compensation, 93.4 % of respondents who suffered an extremely serious injury or illness sought compensation. As with the finding-another-party-responsible-stage, medical treatment-related, work-related, product-related, and motor vehicle-related injuries and illnesses had much higher rates of seeking compensation compared to the overall average.

We also looked at reasons that respondents did not pursue compensation when they believed that another party was responsible to identify additional sources of truncation for the universe of cases that shape the common law. Most of the responses centered around a sense that seeking compensation was something to be generally avoided and that it may not be worth the effort or that they had already received assistance from other sources.

Next, we looked at the rate of contacting an attorney for those who sought compensation. About half contacted an attorney, but interestingly 46 % were turned away by attorneys, typically due to cost of litigation.

We did not collect data on the subsequent winnowing of cases that shape the common law during the litigation process – the decision to appeal the outcome of the trial and the decision of appellate courts to publish a precedential opinion.

Overall, we observe enormous non-random winnowing from the overall universe of potential disputes to the tiny fraction that shape the common-law of torts. What effects does this process have?

First, the threshold perception that another party may be responsible is partly the result of existing societal norms and institutions of claiming and blaming rather than any extra-legal truth of responsibility. In many potential cases, the would-be plaintiff felt that either no one was to blame or that it would not be worth the effort. This is a function of norms and institutions. We are accustomed to blaming others for automobile injuries and other drivers in particular – not typically car manufacturers or road construction authorities. This is partly a result of the path-dependent evolution of our system of compensation for automobile injuries. It is possible that tort law helps shape our collective responsibilities. As we have observed, responsibility or fault is often a prerequisite to pursuing a tort claim. But the causal arrow may point in both directions – tort may help create that sense of fault.

A similar self-reinforcing pattern is true among lawyers’ decision to take a case and file a lawsuit. Certain kinds of cases are worth taking because they are “good” cases. One of the necessary conditions for a case to be worth taking is some precedent for that kind of claim.

Second, we observed and collected data only about the truncation of cases among potential plaintiffs. There are different dynamics of truncation on the defense side. On the one hand, defendants obviously cannot choose which plaintiffs sue them. But because some defendants are repeat-players, insurers, most obviously, they may have more opportunities to choose which cases to settle and which cases to litigate. These effects are outside the scope of our research but likely have important interactions with the effects we observed.