Recent litigation brought against cigarette manufacturers,(1) software companies over potential year 2000 computer problems,(2) and a fast food restaurant for serving coffee that was allegedly too hot(3) reminds us of the importance and dynamic nature of tort law in the United States. Judging from ongoing coverage by newspapers and television, tort law is newsworthy. Yet, as with other legal issues, it is within the covers of law reviews and specialty journals in economics that much of the debate over the social utility of various tort rules and their reform takes place. In that debate law and economics exercises great influence. "Ever since the 1970s, the modern movement in economic analysis has been in full swing. That analysis has highlighted the deterrence function of tort law. Indeed, even in the works of mainstream scholars, deterrence has now assumed the role of a primary rationale for tort liability rules."(4) One example of this influence is the impact of economic analysis of tort law on the revision of the Restatement of Torts (Second) sections on products liability.(5)
In spite of the significance of tort law and the economic analysis of it, the general public, practicing attorneys, and legislators often know little about the findings and informed opinions of those scholars specializing in law and economics. The purpose of this Article is neither to review contemporary issues surrounding tort law(6) nor to gauge the extent of the influence of specialists in law and economics; our purpose is to address whether a consensus exists among these scholars about a few fundamental doctrines of tort law. Because efficiency is a major concern in the field of law and economics, each proposition raises an issue of efficiency about a tort rule. We thus framed ten propositions about how efficiently tort rules achieve their purposes.
In the following section we present our results as a whole. Next we discuss the results individually, offering brief resumes of the debates that inspired the particular questions. Finally, we offer some general conclusions based on the results taken together.
As part of a larger survey conducted during the summer of 1996,(7) we sent questionnaires to members of the American Law and Economics Association (ALEA).(8) We received useable responses to 16.2% of the 389 surveys we mailed out. Respondents were asked to classify themselves as legal scholars (on the basis of their having a law degree), as economists (on the basis of their having a Ph.D. in economics), or as both (on the basis of their having a law degree and a Ph.D. in economics). Next they were asked to note whether they agreed with each of ten propositions about tort law, using a five point scale in which i equals "strongly disagree" and 5 equals "strongly agree."(9) The percentage of those surveyed who responded to each proposition and the total distribution of responses for each proposition are reported. In addition, we calculated the average numerical scores for each of the three groups of scholars: legal scholars (Law), economists (Econ), and those with training in both (Law/Econ), for each proposition. We estimated t-statistics for each statement, to test for statistically significant differences in the mean scores of the three groups, and chi-square tests, to determine if there are statistically significant differences in the distribution of responses for each proposition within each group.(10) The results are presented in the Table.
TABLE Responses(12) 1. No fault automobile insurance is efficient. Strongly Disagree 21% 32% 20% 23% 4% Strongly Agree Percent Responding: 89% Average: 2.55 Law(13): 2.51 Econ(14): 2.67 Law/Econ(15): 2.40 2. Contributory negligence is more efficient at producing optimal behavior than comparative negligence. Strongly Disagree 7% 32% 37% 19% 6% Strongly Agree Percent Responding: 86% Average: 2.83 Law: 2.76 Econ: 3.13 Law/Econ: 2.40 3. Liability rules are inefficient in the face of risk averse behavior. Strongly Disagree 24% 40% 30% 4% 2% Strongly Agree Percent Responding: 79% Average: 2.20 Law: 2.16 Econ: 2.29 Law/Econ: 2.20 4. The standard of care under a negligence rule does not induce an optimal level of activity. Strongly Disagree 11% 38% 23% 15% 12% Strongly Agree Percent Responding: 84% Average: 2.81 Law: 2.90 Econ: 2.33(16) Law/Econ: 3.60 5. Joint and several liability in product liability cases is efficient. Strongly Disagree 18% 33% 20% 27% 2% Strongly Agree Percent Responding: 87% Average: 2.62 Law: 2.72 Econ: 2.29 Law/Econ: 1.80 6. Holding Good Samaritans liable for negligence in rescue attempts would be inefficient. Strongly Disagree 2% 9% 22% 50% 17% Strongly Agree Percent Responding: 86% Average: 3.70 Law: 3.56 Econ: 4.00 Law/Econ: 3.60 7. Product Liability law is more efficiently handled at the state rather than at the federal level. Strongly Disagree 13% 30% 21% 25% 11% Strongly Agree Percent Responding: 84% Average: 2.91 Law: 2.89 Econ: 2.94 Law/Econ: 3.00 8. Limiting punitive damages to three times actual damages would be more efficient than the current procedure. Strongly Disagree 9% 16% 11% 45% 20% Strongly Agree Percent Responding: 89% Average: 3.50 Law: 3.41 Econ: 3.88 Law/Econ: 3.50 9. Allowing injurors to deduct the proceeds of injured victims' insurance from the damages injurors must pay would be efficient. Strongly Disagree 27% 38% 21% 13% 2% Strongly Agree Percent Responding: 89% Average: 2.25 Law: 1.91 Econ: 2.76(17) Law/Econ: 2.80(18) Chi-square=8.74 10. Permitting attorneys to charge contingent fees is inefficient. Strongly Disagree 23% 57% 15% 5% 0% Strongly Agree Percent Responding: 95% Average: 2.02 Law: 2.17 Econ: 1.78 Law/Econ: 1.80
For eight of the ten propositions there are no differences in the responses of legal scholars, economists, and those who are trained in both disciplines. Differences among the groups for two of the propositions, (4 and 9), are discussed below.(11)
NO FAULT AUTO INSURANCE
During the 1960s and 1970s "no-fault" insurance plans, primarily for automobile accidents, became a popular reform of the tort system.(19) Plans with significant "no-fault" components continue to be proposed for automobile accidents(20) and for other areas, such as medical malpractice.(21)
Under a "no-fault" scheme, the injured party is compensated by his insurance company without regard to fault and his ability to bring suit against the tortfeasor is limited.(22) Thus, the distinguishing features of no-fault schemes are: (1) their substitution of first-party insurance for third-party insurance; (2) their limitation of recovery to purely pecuniary losses; and (3) their limitation on the ability to bring tort suits.(23) Existing no-fault systems differ in many details, such as whether they preserve tort claims for injuries above a threshold, but all include these features.(24)
There are two significant economic issues surrounding no-fault schemes. First, by substituting first-party insurance for tort claims, no-fault schemes eliminate the deterrent function of the tort system, emphasizing instead the compensation function.(25) Second, by replacing litigation with insurance payments, no-fault schemes provide (at least) theoretically lower administrative costs.(26) The efficiency of no-fault insurance thus turns on how one compares the efficiency losses imposed by the loss of the deterrent function of the tort system against the gains from lowering administrative costs. Some data exists to assist in this evaluation. Quebec, New Zealand, and various American states have experimented with no-fault schemes for decades,(27) although interpretation of the results of their experiments varies.(28) The standard law and economics analysis of the issue focused on this tradeoff and resolved it in favor of deterrence.(29)
Given the lack of an empirical resolution of the tradeoffs between deterrence and compensation and disagreement over the magnitude of administrative costs, it is not too surprising that there was no strong consensus among respondents about whether "[n]o fault automobile insurance is efficient.(30) Nonetheless, almost twice as many, (53%), disagreed at least somewhat with the proposition as agreed, at least somewhat, (27%), with the proposition.(31) The responses were similar for all three groups--while there are differences of opinions within group classifications, there are no differences in the average opinions across groups.(32) The results thus provide some indication that the deterrent function of the tort system remains relatively important for law and economics scholars compared to the compensation function.
COMPARATIVE VS. CONTRIBUTORY NEGLIGENCE
One of the most sweeping changes in tort law during the twentieth century was the shift from contributory negligence to comparative negligence. "In 1900, the basic rule of tort liability in every state was negligence with a defense of contributory negligence. As of 1986, all but six states and the District of Columbia had switched to a comparative negligence standard.(33) Somewhat surprisingly, this shift "was not preceded by pervasive support in the scholarly literature on tort liability."(34) Almost as striking as the doctrinal shift is the change in the consensus in law and economics scholarship. Until the mid-1980s, most law and economics scholarship found the contributory negligence rule efficient while finding that the comparative negligence rule was not.(35) From the mid-1980s on, comparative negligence enjoyed a stunning growth in support in the scholarly literature. The subject therefore struck us as particularly appropriate for our survey.
Both contributory and comparative negligence are defenses to liability rather than theories of liability. Under contributory negligence, when a plaintiffs negligence contributes to an accident, the plaintiff is barred from recovery.(36) Under comparative negligence, on the other hand, a plaintiffs recovery is reduced but generally not entirely barred by his negligence.(37)
The original case for contributory negligence appeared in a 1973 article by John Prather Brown(38) and received support from informal analyses by Guido Calabresi and Richard Posner.(39) The focus of these analyses was on the optimal sharing of care taking between injurors and tortfeasors.(40) The second wave of law and economics scholarship found that comparative negligence is generally not inefficient compared to contributory negligence, largely because it incorporated more sophisticated understandings of how the comparative negligence rule operates.(41) Importantly, the increased sophistication of the second wave of scholarship is reflected in many cases in the greater degree of qualification of its conclusions--these scholars did not find that comparative negligence is always more efficient, but took steps to identify the conditions under which comparative negligence is efficient relative to contributory negligence? Thus one might find evidence of a "strong view" of the efficiency of either rule (one is always more efficient than the other) or of a "weak view" (circumstances dictate which is more efficient).
For Proposition 2, 39% of respondents disagreed that contributory negligence is more efficient than comparative negligence in producing optimal behavior, while 25% agreed with the proposition.(43) Thirty-seven percent were neutral.(44) These results suggest that a substantial number of law and economics scholars either do not accept the validity of the strong view of the efficiency of either rule, or find that the conditions under which comparative negligence's efficiency advantages exist are not widespread in practice. They also suggest that the lessons of the second wave of scholarship have largely been accepted.
RISK AVERSION AND EFFICIENT TORT RULES
For simplicity's sake, many economic models of tort rules (and indeed economic models generally) include the assumption that the parties are risk …