Inconsistency of Fairness Evaluation in Simulated Labot Market

Reciprocal behavior was often explained by perception of fairness derived from either agents' intention or distributional outcome. In this paper, we demonstrated that fairness perception depended on the evaluability of the partner's type. We conducted experiments to investigate how workers formed fairness perception on the employers. We found inconsistency in fairness evaluation in the two simulated worker-employer relations; workers derived fairness by comparing own wage with market wage in a one shot interaction, but workers derived fairness based on current and previous wage when interacting with same employer. The reversal of fairness perception suggested the role of evaluability of partners' attribute in effort decision among workers