Behavioral IdentiÞcation in Coalitional Bargaining: An Experimental Analysis of Demand Bargaining and Alternating Offers∗

We compare alternating-offer and demand bargaining models of the legislative bargaining process. These two approaches make very different predictions in terms of both ex-ante and ex-post distribution of payoffs, as well as about the role of the order of play. Experiments show that actual bargaining behavior is not as sensitive to the different bargaining rules as the theoretical predictions. We compare our results to studies attempting to distinguish between these two approaches using Þeld data. We Þnd strong similarities between the experimental data and the Þeld data regardless of whether the experiments employ alternating-offer or demand-bargaining protocols. This behavioral identiÞcation problem suggests that it is impossible to derive, just from payoff data, what bargaining rules are being used in coalitional bargaining outside the laboratory. ∗Morelli’s research was partially supported by National Science Foundation Grant No. 0213312. Kagel’s research was partially supported by National Science Foundation Grants No. 0136925 and 0136928 and a grant from the Mershon Center at the Ohio State University. Any opinions, Þndings, and conclusions or recommendations in this material are those of the authors and do not necessarily reßect the views of the National Science Foundation or the Mershon Center. We have beneÞtted from helpful discussions with James Peck, Maria Montero, and participants at the Experiments III session of the 2003 North American Summer Meeting of the Econometric Society, the Stony Brook Workshop on Experimental Game Theory, and the NOM Seminar at the Harvard Business School.