Public bad prevention by majority voting on redistribution - Experimental evidence

We present an experimental study on voting behavior in groups of seven persons where public bad prevention depends on redistributing income by qualified majority voting. Although a payoff-maximizing voting pattern exists which guarantees a qualified majority thus maximizing both individual and group payoffs the quali-fied majority is failed in 27% (relatively costly public bad), respectively 46% ( relatively cheap public bad) of all decisions. Controlling for different degrees of social distance when casting votes (i.e. anonymous versus with identification) we find that social distance matters when stakes are relatively low, inducing less efficient outcomes for the group when voting behavior is revealed. The endogenously determined status of subjects with respect to redistribution (unlucky subjects lose, lucky subjects win) systematically influences subjects voting strategies and the collective outcomes, such that a higher number of unlucky subjects leads to less efficient outcomes for the group.