In this paper we report on a principal-agent experiment where the principal can choose whether to rely on an unenforcable bonus contract or to combine the bonus contract with a fine if the agent's effort falls below a minimum standard. We show that most principals do not use the fine and that the pure bonus contract is more efficient than the combined contract. Our experiment suggests that principals who are less fair are more likely to choose a combined contract and less likely to actually pay the announced bonus. This offers a new explanation for why explicit and implicit incentives are substitutes rather than complements.
moral hazard bonus contract implicit incentives fairness incentives