مذاکره قبل ازجدال در پشت درهای باز در مقابل درهای بسته : تجزیه و تحلیل اقتصادی از قانون 408

In this paper, we examine the economics of Rule 408 of the Federal Rules of Evidence, whereby the negotiation processes at the pretrial stage are made inadmissible to prove the amount of liability. It is asserted that under Rule 408 a settlement offer is less inclined to be rejected since the judge is expected to make a lower award given the signal he observes. Also, we derive a sufficient condition for Rule 408 promoting settlement.

The litigiousness of American society has been a growing social concern. Time-series data show that, over the course of the 20th century, there is an upward trend in civil cases filed as percentage of the population in many American jurisdictions.1 This has recently drawn the attention of economists as well as legal scholars into the area of litigation process. Many legal devices have been designed to reduce court congestion and legal expenditures. These devices include fee-shifting rules such as Rule 68 of the Federal Rules of Civil Procedure2 discovery requirements3 a shift in a certain tort from a rule of negligence to a rule of strict liability4 and Rule 408 of the Federal Rules of Evidence.5 The purpose of this paper is to examine whether Rule 408 of the Federal Rules of Evidence can promote settlement and thus reduce the number of actual trials. The idea of protection against the admissibility of conducts made in compromise negotiations goes back prior to the adoption of the Federal Rule of Evidence in 1975. The rationale relied upon by courts at common law in excluding evidence of the actual compromise offer was that they were irrelevant to the substantive issues; such conduct implied merely a desire for peace, not a concession of weakness of position. Another rationale was that it was improving social welfare to encourage the out-of-court resolution of disputes (Brazil, 1988). To promote this purpose, the Advisory Committee, in drafting Rule 408, expanded its range of protection considerably; specifically, the rule offered protection not just to offers or demands, but also to conduct or statements made in compromise negotiations. Economic analysis of Rule 408 was first attempted by Daughety & Reinganum (1995). By analyzing a model where the informed plaintiff makes a take-it-or-leave-it offer, they show that making a pretrial demand admissible will increase the expected number of cases that go to trial, since the defendant must reject the plaintiff’s offer more often in order to counteract the plaintiff’s incentive to inflate their demand with intent to influence the judge’s beliefs under admissibility. However, we think that their analysis of Rule 408 is partial, in the sense that it only involves the effect of protection against the admissibility of pretrial demands. The analysis would be completed when one considers the effect of protection against admissibility of other kinds of conduct made in compromise negotiations as well. In this paper, we will show that Rule 408 can increase the rate of settlement by making inadmissible the pretrial evidence that settlement offer of a certain amount has been rejected by the plaintiff, in an alternative model where the uninformed defendant makes a take-it-or-leave-it offer.6 The intuitive reason is that a plaintiff would be more likely to reject the defendant’s settlement offer to influence the judge’s belief on his damage amount upward and to make his award at court higher without Rule 408. Considering this alternative model will be especially important and worthwhile when a policy maker is preparing a similar rule protecting the confidentiality of pretrial negotiations. The analysis offers useful policy implications regarding the extent to which the protection should be given. Our model has much in common with that of Daughety & Reinganum (1995) but is distinguished in several respects. In their model, the judge could elicit useful information on the plaintiff’s actual damage amount from the pretrial settlement demand of the plaintiff, while in our model the judge can update his information on the plaintiff’s damage amount by using the fact that a settlement offer of a certain amount has been rejected. Also, Daughety & Reinganum (1995) assume that the trial reveals either perfect information or no information at all to the judge, while we assume that the judge can discover some noisy additional information during the trial. Consequently, their model exhibits the feature that, if there is a very high likelihood that the trial reveals no information, all the cases go to trial under admissibility. By contrast, in our model, such an unrealistic feature does not appear. The rest of the paper is organized as follows. In Section 2, we set up the model. We analyze the case in which pretrial negotiations are not admissible as evidence in Section 3, and the case in which pretrial negotiations are admissible in Section 4. Comparison of the two cases is made in 5 and 6 contains discussion and concluding remarks.

In this paper, we have shown that Rule 408 can promote settlement by making conduct in pretrial settlement negotiations inadmissible in court, leading to less litigation. Our work addresses a similar issue as that of Daughety & Reinganum (1995), but differs from theirs on two fronts. First, we consider a model in which the uninformed defendant makes a take-it-or-leave-it settlement offer to the informed plaintiff, contrary to their model in which the informed plaintiff is the first mover. By adopting such an alternative model, we demonstrated the signaling effect of rejecting an offer at the pretrial stage on the settlement rate under admissibility, while Daughety & Reinganum (1995) showed the signaling effect of the pretrial settlement demand. Second, the court error is modeled in a different fashion in the two models. In our model, the judge is assumed to receive a continuously distributed signal, whereas, in their model, the judge is assumed to be either completely informed of the true value or absolutely ignorant. Since their assumption of binary information revelation is rather restrictive compared with our continuous distributional assumption, our interpretation of the court error seems more natural. In view of these differences, our model complements and improves on Daughety & Reinganum (1995). Our model can be extended in various directions. We may consider a model of two-sided uncertainty in which both the defendant and the plaintiff have private information of their own. But, the analysis of such a model appears quite complicated and lies beyond the scope of this paper.9 Also, it may be more realistic to assume that settlement offers are made not on a take-it-or-leave-it basis, but take place over periods. We believe, however, that the extended feature will not change the qualitative nature of the results in this paper, since, in many dynamic settlement models, the settlement procedure does not last for more than one period in equilibrium.10 Finally, it would be possible to consider a larger game in which the defendant’s ex ante care level is incorporated explicitly into the model. This is indeed an interesting extension, but it also widens the scope of the paper significantly. So, it will be left for future research. , , , , and