Alternative institutions for resolving coordination problems: Experimental evidence on forward induction and preplay communication. (English) Zbl 0874.90208
Friedman, James W. (ed.), Problems of coordination in economic activity. Boston, MA: Kluwer Academic Publishers. Recent Economic Thought Series. 35, 129-146 (1994).
The point of these experiments has been to provide an understanding of the nature of coordination problems endemic to many economic interactions. In our experimental work on coordination, we have established that agents do not play the Pareto-dominant Nash equilibrium in a variety of coordination games and that ex post disequilibrium arises quite often in the battle of the sexes game.
In light of these coordination difficulties, it is quite natural to consider alternative institutions (i.e., different extensive form games) that might overcome them. We summarized the results from two important variations on these games: the presence of an outside option and pre-play communication. These institutions are both relevant in that they provide a means of coordinating activities. Moreover, these same institutions have received considerable attention in the game theory literature.
Our findings indicate that neither preplay communication nor the presence of outside opportunities will necessarily solve all coordination problems. In particular, we find that the nature of the strategic interaction between players – coordination v. conflict – determines which institution fares better. Our work indicates that in games of conflict, such as the BOS game, one-way communication or an outside option for one agent is sufficient for resolving ex post disequilibrium problems because these institutions provide an asymmetry in the game. In fact, we find that even a payoff irrelevant outside option influences play through the creation of an asymmetry.
For coordination games, in contrast, the issue is confidence and the creation of an asymmetry is not relevant. For CG, we find that only through two-way communication is it possible to build enough confidence to attain the Pareto-dominant Nash equilibrium.
These results are suggestive along two lines. First, they point to the need to develop a crisper characterization of the differences between games of conflict and games of coordination. Second, using this characterization, one should continue to test the theme of this paper regarding the nature of the strategic interaction and the choice of institution for resolving the particular coordination problem.
For the entire collection see [Zbl 0861.00029].
In light of these coordination difficulties, it is quite natural to consider alternative institutions (i.e., different extensive form games) that might overcome them. We summarized the results from two important variations on these games: the presence of an outside option and pre-play communication. These institutions are both relevant in that they provide a means of coordinating activities. Moreover, these same institutions have received considerable attention in the game theory literature.
Our findings indicate that neither preplay communication nor the presence of outside opportunities will necessarily solve all coordination problems. In particular, we find that the nature of the strategic interaction between players – coordination v. conflict – determines which institution fares better. Our work indicates that in games of conflict, such as the BOS game, one-way communication or an outside option for one agent is sufficient for resolving ex post disequilibrium problems because these institutions provide an asymmetry in the game. In fact, we find that even a payoff irrelevant outside option influences play through the creation of an asymmetry.
For coordination games, in contrast, the issue is confidence and the creation of an asymmetry is not relevant. For CG, we find that only through two-way communication is it possible to build enough confidence to attain the Pareto-dominant Nash equilibrium.
These results are suggestive along two lines. First, they point to the need to develop a crisper characterization of the differences between games of conflict and games of coordination. Second, using this characterization, one should continue to test the theme of this paper regarding the nature of the strategic interaction and the choice of institution for resolving the particular coordination problem.
For the entire collection see [Zbl 0861.00029].
