The dynamics of individual preferences in strategic and non-strategic games
Weimann, J.*, Sass, M.*, Timme, F.*
Duration: 2012 - 2015
The dominant behavioral theory in economics presupposes that people have complete preference orders that can be represented by utility functions. Under this assumption, decisions can be described as benefit maximization calculations. The basis for this is the theory of revealed preferences, which ensures that the rational model can be applied under very general conditions. The theory of revealed preferences says nothing about whether preference orders are constant over time or can change through election decisions. However, the application of the rational model to the analysis of individual decisions in economics implicitly presupposes that preferences remain constant over time. Only under this assumption can equilibria be compared and the method of comparative statics be applied. Whether preferences are actually constant is ultimately an empirical question, but one that is difficult to answer with field data. This is why experimental methods are appropriate. To date, there are only a few findings that can be traced back to the applicant's preliminary work. These show that social preferences in particular are not stable, but show a clear dynamic: Even without learning effects and without the possibility of acquiring a reputation, the influence of social preferences is considerably reduced in repeated dictator experiments. The project will examine whether this finding also applies to strategic games. For this purpose, an experimental design will be used that excludes learning and reputation effects. First results of a pilot experiment have shown that the preferences of the players in public goods experiments are subject to a similar dynamic as in dictator experiments. The design will be used in other strategic and non-strategic games and the question of how preferences change in the long run will be clarified. The basic hypothesis is that preferences change through adaptation processes, but lose their importance during long pauses between decisions, so that (social) preferences can regenerate in the long run. In addition, the project will deal with the question of which adaptation processes drive behavioral dynamics.
* Otto-von-Guericke-Universität Magdeburg
Funding: DFG