Litigation and settlement in a game with incomplete information (Q1921012)

The book develops some ideas on learning in a two-person bargaining game with asymmetric information where subjects -- the defendant and the plaintiff have the chance of settling the conflict out-of-court. The author uses Schweizer's game with some modifications as a guideline to the experiment [see \textit{U. Schweizer}, Rev. Econ. Stud. 56, No. 2, 163-177 (1989; Zbl 0672.90029)]. In the game both players decide simultaneously after the defendant makes two settlement proposals and the plaintiff determines his acceptance limit. After both players have made their choices and nature has chosen the defendant's position, the respective settlement offer is compared with the acceptance limit \(A\). For any offer below \(A\) the case will be litigated and a lottery decides on the outcome. Two learning processes have been found in the experimental investigation of this game. The first process considers the way of how subjects adjust their values. The other learning process refers to the polarization of acceptance limits when subjects repeat the game for the first time. It is concluded that the adjustment of values is strongly guided by the ex-post bargaining assessment. Dependent upon the bargaining outcome, subjects think about how they could have improved their action.