Self-selecting random or cumulative pay? A bargaining experiment

An incentive scheme in a multi-task experiment may trigger different types of behavior in participants. This experimental study allows participants to decide between being paid to complete a random task or being paid an average across all tasks completed in a bargaining game where the buyer and seller have partly conflicting interests and are asymmetrically informed. We find that both past experience and individual characteristics have a significant effect on the payment scheme selected and the final trading result. Although the payment method is likely to be correlated with risk preferences, it may also be influenced by ego-defensive concerns. Analysis of the self-selected Random Lottery Incentives scheme (RLI) and the Cumulative Scheme (CS) shows both individual and social effects: sellers who prefer CS over RLI receive a larger share of the surplus, and (female) buyers choosing RLI play with a lower degree of trust with the aim of closing better deals.