Conflict, Cooperation and Coordination : Essays in Game Theory and Experimental Economics

Sammanfattning: This thesis consists of four independent papers from two categories. The first two papers are theoretical whereas the other two report results from laboratory experiments. What unites all four papers is that they are concerned with game theory, i.e. the formal analysis of strategic interaction. The core concept in game theory is that of Nash equilibrium, which describes a situation in which, given everyone else’s behavior, I cannot improve my outcome by changing my own behavior, and the same holds for all other parties involved. On the theoretical side, in one paper we apply level-k reasoning to contests whereas in the other we derive a Folk Theorem for infinitely repeated game protocols when players’ preferences are modulated by endogenous reference points. While the two theoretical papers are concerned with deriving precise results for classes of games, the experimental papers investigate how people actually behave in situations where game theory does notyield precise predictions. In other words, they address open questions in equilibrium selection. Under study are (a) a bargaining problem reminiscent of worker-firm interaction in real-world labor markets, and (b) a coordination problem with heterogeneous agents in which mis-coordination is very costly (think of the question of funding the EFSF and ESM, with the aim of erecting a “firewall” for troubled eurozone countries, as an example). In problem (a) our question is how market frictions arising from institutional factors (e.g. employment protection) and market conditions (e.g. scarcity of skilled labor) affect efficiency and surplus sharing between workers and firms. We find that while efficiency is significantly lower the higher the amount of frictions, division of surplus is mostly equitable in functioning matches. In problem (b) we ask to which extent heterogeneity with respect to observable characteristics (e.g. wealth) causes coordination frictions, and to which extent these frictions can be attributed to proper normative disagreement, or to a multiplicity of focal points and resulting strategic uncertainty. We find that the extent of frictions is substantial but mostly due to strategic uncertainty. Intrinsic preferences seem to play a relatively minor role even though the revealed focal points reflect basic fairness principles.