Learning, Non-equilibrium Beliefs, and Non-pecuniary Payoffs in an Experimental Game

We present and estimate a parametric learning model of players' dynamic and possible out-of-equilibrium beliefs about other players'social preferences using the data from the four-country ultimatum game experiments of Roth et al. (1991). The model allows for each of the three leading factors that have been considered in the literature on these games: random utility, non-pecuniary preferences, and learning. We find evidence that in the US and in Israel, the estimated beliefs of proposers are stationary and out-of-equilibrium, that in Slovenia, they are in equilibrium, and that in Japan, they change from period to period and move away from equilibrium over time. On average, proposers and responders have negative regard for each other's monetary payoffs in all countries.