09 October 2018
Abstract
Zeuthen Hicks bargaining is a dynamic bargaining model that, in contrast to other models, allows to make testable predictions about which side in a negotiation makes a concession, and the extent of these concessions. According to the model, negotiators make bargaining steps that increase the value of the product of both parties' utilities, i.e. the function that is also maximized in the Nash bargaining solution. Thus, literature argues that negotiators following the model will end up at the Nash bargaining solution. In the present paper, we show that this is not necessarily the case. Depending on the shape of utility functions of the negotiators, the function maximized might have several local optima, so the process can fail to reach the Nash bargaining solution. We study conditions under which such multiple local optima exists both analytically an in a computational study. Furthermore, we test, using data from experiments with the electronic negotiation support system Inspire, whether negotiators in fact act according to the predictions of the model, and whether negotiators who more frequently adhere to the model achieve better outcomes. Empirical result indicate that typically, concessions fall short of what is required by the model, but that adhering to the model's predictions would in fact lead to better bargaining outcomes.
International Institute for Applied Systems Analysis (IIASA)
Schlossplatz 1, A-2361 Laxenburg, Austria
Phone: (+43 2236) 807 0 Fax:(+43 2236) 71 313